LJBRARJES remove this checkout from
Jag-[gan-L. your record. FINES wilI

, be charged if book is
returned after the date
stamped be1ow.

MSU ' RETURNING MATERIALS:
Place in book drop to

 

 

 

 

 

 

 

 

MECHANISMS FOR THE RELEASE OF IRON FROM
FERRITIN AND THEIR RELATIONSHIP TO LIPID PEROXIDATION
AND TOXICITY

BY

Craig Eugene Thomas

A DISSERTATION
Submitted to
Michigan State University
in partial fulfillment of the requirements

for the degree of

DOCTOR OF PHILOSOPHY

Department of Biochemistry

1986

2./,/6‘0*2

ABSTRACT
MECHANISMS FOR THE RELEASE OF IRON FROM FERRITIN
AND THEIR RELATIONSHIP TO LIPID PEROXIDATION
AND TOXICITY
BY

CRAIG EUGENE THOMAS

This work was undertaken in an effort to determine what
physiological iron chelates may provide iron for redox reactions
leading to the formation of oxidants which initiate membrane lipid
peroxidation (LP). Initial studies were designed to separate and
characterize low molecular weight (MW) iron complexes from rat liver
cytosol. Gel filtration experiments revealed that essentially 100% of
the total cytosolic iron was recovered in fractions ranging from
100-500,000 MW, with 70% of the iron present in ferritin, the major
iron storage protein.

Subsequent work demonstrated that superoxide (027), generated by
xanthine oxidase, released iron from ferritin and initiated the
peroxidation of phospholipid liposomes. Iron release and LP were
inhibited by superoxide dismutase (SOD) but stimulated by catalase.
ESR spin trapping studies supported an inverse relationship between
hydroxyl radical formation and LP. Catalase stimulated LP by
preventing rapid HZOZ-dependent oxidation of Fe2*, allowing more
efficient formation of an initiator of LP.

Superoxide, generated by the redox cycling of paraquat and
catalyzed by NADPH-cytochrome PMSO reductase, also released iron from

ferritin and promoted LP. SOD inhibited paraquat~dependent Fez+

Craig Eugene Thomas

release only 50-60%. Anaerobically, the paraquat cation radical
rapidly released all of the iron from ferritin, accounting for the
SOD-insensitive Fe2* release.

The antitumor anthracycline antibiotics, adriamycin and
daunomycin, whose clinical use is limited by cardiotoxicity, also
underwent redox cycling to generate 027, releasing Fez+ from ferritin.
ESR studies demonstrated that, anaerobically, their semiquinone free
radicals transferred electrons to ferritin to release Fe2+.
Accordingly, aerobic iron release was inhibited only 50% by SOD. The
semiquinone radical of diaziquone, a relatively non-toxic
chemotherapeutic drug, was incapable of electron transfer and Fez+
release.

Another study characterized the endogenous iron in rat liver
microsomes. Deve10pment of an ELISA for ferritin revealed that 83% of
the iron in microsomes is in ferritin. A chromatographic procedure
was devised which removed ferritin, as well as contaminating SOD and
catalase, from microsomes. The iron in ferritin was released by 027,
whereas the remaining 17% of the iron was directly reduced by the

reductase. Both iron pools were capable of supporting microsomal LP.

To My Grandmother and Grandfather (late)
Ma-Mu and Pap-Pap

ii

ACKNOWLEDGEMENTS

The author wishes to express his sincere appreciation to Dr. Steven
D. Aust. His enthusiasm, patience, and guidance were invaluable to
this dissertation and the development of my research interests. I
would also like to thank the other members of my guidance committee:
Drs. Ian Gray, William Smith, Clarence Suelter, and William Hells for

their very informative and helpful discussions.

The support of all of the laboratory was also essential to getting
through the sometimes ”trying" experience of graduate school. The good
times, both in and out of the laboratory, will always be remembered.
Special thanks must go to Lee Morehouse, with whom I worked most
closely, and who helped to convince me that there are universities

other than Penn State.

Much of the credit for this work must go to my family, particularly
my parents. It is they who subtilely impressed upon me the value of

education and the virtues of hard work.

Finally, I cannot thank enough my wife, Jean, who saw me through the
difficult times and shared in the laughter of the good times,and whom
had the pleasure (?) of typing this dissertation. Her continued

support made it all worthwhile.

iii

TABLE OF CONTENTS

LIST OF TABLES . . . . . . . . . . . . . . . . . . . . . . .
LIST OF FIGURES . . . . . . . . . . . . . . . . . . . . . .
LIST OF ABBREVIATIONS. . . . . . . . . . . . . . . . . . . .
INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . . .
LITERATURE REVIEW. . . . . . . . . . . . . . . . . . . . . .

Univalent Reduction of Molecular Oxygen . . . . . . . . .

1. Superoxide and Hydrogen Peroxide. . . . . . . . . .

2. Formation of the Hydroxyl Radical . . . . . . . . .

Lipid Peroxidation. . . . . . . . . . . . . . . . . . . .
1. An Overview of Initiation, Propagation and

Termination. . . . . . . . . . . . . . . . . . .

2. Initiation by the Hydroxyl Radical. . . . . . . . .

3. Alternatives to the Hydroxyl Radical in Initiation.

Iron Metabolism . . . . . . . . . . . . . . . . . . . . .

1. Absorption and its Regulation . . . . . . . . . . .

2. Transport of Iron . . . . . . . . . . . . . . . . .

3. Iron Storage in Ferritin: Uptake and Release . . .

A. Biological Iron and Lipid Peroxidation. . . . . . .

Iron in Proposed Oxygen Radical—Dependent Toxicities. . .

1. Paraquat. . . . . . . . . . . . . . . . . . . . . .

2. Adriamycin. . . . . . . . . . . . . . . . . . . . .

LIST OF REFERENCES . . . . . . . . . . . . . . . . . . . . .
CHAPTER
I. FERRITIN AND SUPEROXIDE-DEPENDENT LIPID PEROXIDATION .
Abstract . . . . . . . . . . . . . . . . . . . . . . .
Introduction . . . . . . . . . . . . . . . . . . . . .

Materials and Methods. . . . . . . . . . . . . . . . .
' Materials . . . . . . . . . . . . . . . . . . . . .
Enzymes . . . . . . . . . . . . . . . . . . . . . .
Fractionation of Ra Live Cytosol. . . . . . . . .
Purification of Rat Liver Ferritin. . . . . . . . .
Preparation of Microsomal Lipid and Liposom s . . .
Lipid Peroxidation Assays . . . . . . . . . . . . .
Assays for Total Iron and Ferritin Iron Release . .
ESR Spin Trapping . . . . . . . . . . . . . . . . .

iv

Page

viii
ix

xi

Page
A6

A7
”8

52
52
52
53
53
5A
55
55
56

CHAPTER

II.

III.

ReSUItS O O O O O O O I O O O O O O I O O O O O O O O

Fractionation of Rat Liver Cytosol. . . . . . . . .
Xanthine Oxidase—Dependent Lipid Peroxidation . . .
Effect of SOD and Mannitol. . . . . . . . . . . . .
Release of Iron from Ferritin . . . . . . . . . . .
ESR Spin Trapping . . . . . . . . . . . . . . . . .

Discussion . . . . . . . . . . . . . . . . . . . . . .
List of References . . t . . . . . . . . . . . . . . .
PARAQUAT AND FERRITIN-DEPENDENT LIPID PEROXIDATION . .
Abstract . . . . .2. . . . . . . . . . . . . . . . . .
Introduction . . . . . . . . . . . . . . . . . . . . .

Materials and Methods. . . . . . . . . . . . . . . . .
Materials . . . . . . . . . . . . . . . . . . . . .
Enzymes . . . . . . . . . . . . . . . . . . . . . .
Preparation of Ferritin and Assay for Total Iron. .
Preparation of Microsomal Lipid and Liposomes . . .
Lipid Peroxidation Assays . . . . . . . . . . . . .

Results. . . . . . . . . . . . . . . . . . . . . . . .
Paraquat and Ferritin-Dependent Lipid
Peroxidation . . . . . . . . . . . . . . . . . .
Effect of SOD, Catalase, and Mannitol on Paraquat
and Ferritin-Dependent Lipid Peroxidation. . . .
Paraquat and Ferritin-Dependent Lipid Peroxidation
in the Absence of ADP. . . . . . . . . . . . . .

DASCUSSAOH O O O O O O I O 0 O O O O O O O O O 0 O O 0
List of References . . . . . . . . . . . . . . . . . .

REDUCTIVE RELEASE OF IRON FROM FERRITIN BY CATION FREE
RADICALS OF PARAQUAT AND OTHER BIPYRIDYLS . . . . .

Abstract 0 O O O O O O O O O O O O O 0 O 0 0 O O O O O
IntrOdUCtlon O O O O O O O C O O .0 O O O O O O O I O 0
Materials and Methods. . . . . . . . . . . . . . . . .

Materials . . . . . . . . . . . . . . . . . . . . .

Enzymes I O I O O O l O O O O O O O O O O O O O O 0
Preparation of Ferritin and Iron Release Assays. .

Analysis of the Structure and Iron Uptake Ability of'

Ferritin Following Iron Release by Paraquat. . .
Paraquat Radical Detection. . . . . ... . . . . . .

Page

57
57
59
6M

67
67
77 .
83
87
88
89
91
91
91
92
92
92
9H
9H
‘9“
99
111

115

119
120
121
125
125
125
126

127
128

CHAPTER

IV.

Resu1t3 O O O O O O O O O O O I O O I O O O O O O O C O

Page
130

Effect of SOD and 02 on Iron Release from Ferritin . . . 130

Effect of Varying Concentrations of Ferritin, NADPHé

Cytochrome PASO Reductase, and Paraquat on Iron
Release from Ferritin by Paraquat. . . . . . . .
Release of Ferritin Iron and NADPH Oxidation by
Other Bipyridyls . . . . . . . . . . . . . . . .
Effect of Ferritin on Detection of the Paraquat
Radical. . . . . . . . . . . . . . . . . . . . .
Structure and Iron Uptake Ability of Ferritin
Following Iron Release . . . . . . . . . . . . .

Discusgion I O I O O O O O O O O O O O O O O O O O O 0

List of References . . . . . . . . . . . . . . . . . .

RELEASE OF IRON FROM FERRITIN BY CARDIOTOXIC ANTHRACYCLINE

ANTIBIOTICS O O O O O I O O O O O O O O O I O I O O
AbatraCt O 0 O O O I O O O O O O O O I O O O O O O O 0
Introduction . . . . . . . . . . . . . . . . . . . . .

Materials and Methods. . . . . . . . . . . . . . . . .
Materials .‘. . . . . . . . . . . . . . . . . . . .
Enzymes . . . . . . . . . . . . . . . . . . . . .
Preparation of Ferritin and Iron Release Assays . .
ESR Measurements. . . . . . . . . . . . . . . . . .

ReSUJ-ts O O I O O O O O O O O O O O O O O O O O O O O O
Aerobic and Anaerobic Release of Iron from Ferritin

Effect of Ferritin on ESR Detection of Free Radicals.

Discussion . . . . . . . . . . . . . . . . . . . . . .
List of References . . . . . . . . . . . . . . . . . .

RAT LIVER MICROSOMAL NADPH*DEPENDENT RELEASE OF IRON
FROM FERRITIN AND LIPID PEROXIDATION. . . . . . . .

Abstract I O O O O O O O O O O O O O O O O O O O I O O
IntFOduction O O O O O O O O O O O O O O O O O O O O C

Materials and Methods. . . . . . . . . . . . . . . . .
Materials .'. . . . . . . . . . . . . . . . . . . .
Enzymes . . . . . . . . . . . . . . . . . . . . . .
Preparation of Antibodies and Ferritin Quantitation
Preparation of Microsomes . . . . . . . . . . . . .
Lipid Peroxidation Assays . . . . . . .‘. . . . . .
Assays for Total Iron and Ferritin Iron Release . .

vi

3»

132
135
1H0
1U3
151

158

163
16“
165

167
167
167
168
169

170
170
179

187

190

19”
195
~196

198
198
198
199
200
200
201

CHAPTER Page

Results. . . . . . . . . . . . . . . . . . . . . . . . . . 202
Chromatography of Microsomes and Activities of
Associated Enzymes . . . . . . . . . . . . . . . . . 202
Ferritin Iron Release . . . . . . . . . . . . . . . . . 210
NADPH-Dependent Lipid Peroxidation Using '
Purified Ferritin . . . . . . . . . . . . . . . . . 215

DISCUSS1on O O O O O O O O O O I O O 0 O O O O O O O O O O 223
List of References . . . . . . . . . . . . . . . . . . . . 228

APPENDIX 0 O O O O O O I O O O O O O O O O O O O O O O O O O O O 232

vii

LIST OF TABLES

TABLE Page
CHAPTER I

1. Total Iron Analysis of Pooled Fractions Obtained

‘ From Chromatography of Rat Liver Cytosol . . . . . . 58

2. Total Iron Analysis of Rat Liver Homogenate and '
Ultrafiltrated Cytosol . . . . . . . . . . . . . 57

3. The Effect of SOD and Mannitol on Xanthine Oxidase-
Dependent Peroxidation of Phospholipids. . . . . . . 6”

CHAPTER III

1. Effect of SOD on Xanthine Oxidase and Paraquat-

Dependent Release of Iron from Ferritin. . . . . . . 130
2. Effect of SOD on Aerobic, Paraquat-Dependent Release

of Iron from Ferritin. . . . . . . . . . . . . . . . 131
3. Anaerobic Iron Release from Ferritin by Paraquat at

Varying Ferritin Concentrations. . . . . . . . . . . 135
A. Release of Iron from Ferritin and NADPH Oxidation by

Redox Active Bipyridyls. . . . . . . . . . . . . . . 1A0

CHAPTER IV
1. Effect of SOD on Aerobic Release of Iron frdm
Ferritin by Xanthine Oxidase and the Redox
Cycling of Antitumorigenic Agents. . . . . . . . . . 170
2. Anaerobic Release of Iron from Ferritin and NADPH '
Oxidation by Antitumorigenic Agents . . . . . . . . 171

CHAPTER V
1. Effect of Sepharose CL-ZB Chromatography on
Ferritin, Total Iron, and Enzymes Associated
with Microsomes. . . . . . . . . . . . . . . . . . . 209
2. Effect of SOD, CN', and Paraquat on Ferritin Iron -
Release by Microsomes. . . . . . . . . . . . . . . . 218
3. Effect of Catalase, SOD and Paraquat on NADPH- ‘ ‘
Dependent Microsomal Lipid Peroxidation. . . . . . . 219

viii

'Figure

LIST OF FIGURES

CHAPTER I

1.

Ferritin and Xanthine Oxidase-Dependent Peroxi-
dation of Phospholipids. . . . . . . . . . . . . .

The Effect of Varying Catalase Concentrations on
Ferritin and Xanthine Oxidase-Dependent Peroxi-
dation of Phospholipids. . . . . . . . . . .

.The Effect of Catalase on ADP-Fe3+ and Xanthine

Oxidase-Dependent Peroxidation of Phospholipids. .
The Effect SOD and Catalase on Xanthine Oxidase-
Dependent Iron Release from Ferritin . . . . . . .
The Effect of SOD and Catalase on Potassium
Superoxide~Dependent Iron Release from Ferritin. .
ESR Spin Trapping of Superoxide Generated by Xanthine
Oxidase. . . . . . . . . . . . . . . . . . . . . .
ESR Spin Trapping of Hydroxyl Radical Generated by
Xanthine Oxidase and Rat Liver Ferritin. . . . . .

CHAPTER II

1.

The Effect of Varying NADPH-Cytochrome PMSO Reduc-
tase Activity on Paraquat and Ferritin—Dependent
Peroxidation of Phospholipids in the Presence or
Absence of Catalase. . . . . . . . . . . . . .

The Effect of Varying Paraquat (A) or Ferritin (B)
Concentration on Peroxidation of Phospholipids . .

The Effect of Varying Catalase Activity on Peroxida-
tion of Phospholipids Catalyzed by Paraquat and
Ferritin in the Presence of ADP. . . . . . . . .

The Effect of Varying Mannitol Concentration on
Paraquat and Ferritin-Dependent Peroxidation of
Phospholipids in the Presence of ADP and Absence
of Catalase. . . . . . . . . . . . . . . . . . . .

The Effect of Varying ADP Concentration on Paraquat
and Ferritin-Dependent Peroxidation of Phospho-
lipids . . . . . . . . . . . . . . . . . . . . . .

The Effect of Varying Catalase Activity on Paraquat
and Ferritin-Dependent Peroxidation of Phospho-
lipids in the Absence of ADP . . . . . . . . . . .

The Effect of Varying Mannitol Concentration on Para*

quat and Ferritin-Dependent Peroxidation of Phospho-

lipids in the Presence or Absence of Catalase. . .

CHAPTER III

1.

2.

Effect of Oxygen on Release of Iron from Ferritin
by Paraquat. . . . . . . . . . . . . . . . . . . .
Effect of Varying NADPH-Cytochrome PASO Reductase
Activity on the Time Required for Complete Re-
lease of Iron from Ferritin by Paraquat. . . . . .

ix

60

62
65
68
7O
73
75

95

97

100

102
10"
106

108

133

136

Figure Page

3. Effect of Varying Paraquat Concentration on the

Time Required for Complete Release of Iron From

Ferritin by Paraquat . . . . . . . . . . . . . . . . 138
A. Effect of Ferritin on SpectrOphotometric Detection ‘

of the Paraquat Cation Radical . . . . . . . . . . . 1M1
5. Effect of Ferritin on Detection of the Paraquat

Cation Radical by ESR. . . . . . . . . . . . . . . . 1A“
6. Repetitive Scan of the Effect of Ferritin on '

Paraquat Cation Radical Detection by ESR . . . . . . 1A6
7. Iron Uptake by Ferritin . . . . . . . . . . . . . . . . 1A8

CHAPTER IV
1. Effect of Varying Ferritin Concentration on the
' Anaerobic Release of Iron from Ferritin by
Adriamycin . . . . . . . . . . . . . . . . . . . . . 173
2. Effect of Varying Adriamycin Concentration on the '
Anaerobic Release of Iron from Ferritin by
Adriamycin . . . . . . . . . . . . . . . . . . . . . 175
3. Effect of Varying NADPH—Cytochrome PASO Reductase '
Activity on the Anaerobic Release of Iron from
_ Ferritin by Adriamycin . . . . . . . . . . . . . . . 177
A. Repetitive Scan of the Effect of Ferritin on Adria-
mycin Semiquinone Radical Detection by ESR . . . . . 181
5. Repetitive Scan of the Effect of Ferritin on Dauno-
mycin Semiquinone Radical Detection by ESR . . . . . 183
6. Repetitive Scan of the Effect of Ferritin on Diazi*
quone Semiquinone Radical Detection by ESR . . . . . 185

CHAPTER V
1.A Protein Elution Profile for the Chromatography
' of Microsomes and Purified Rat Liver Ferritin
on Sepharose CL- 28 . . . . . . . . . . . . . . . . . 203
1.8 Ouchterlony Double Diffusion Analysis of Microsomes . . 205
'2. ELISA Standard Curve. . . . . . . . . . . . . . . . . . 207
3. Effect of Varying Paraquat Concentration on NADPH- '
' Dependent Iron Release from Ferritin . . . . . . . . 211
A. Effect of Varying Microsomal Protein Concentration
on NADPH-Dependent Iron Release from Ferritin. . . . 213
5. Effect of Varying Ferritin Concentration on NADPH- ‘
Dependent Iron Release from Ferritin . . . . . . . . 216
6. EDTA Titration of NADPH-Dependent Microsomal Lipid '
Peroxidation . . . . . . . . . . . . . . . . . . . . 221

ADP

BCA

DDC
DMPO
DMPO'OH
DMPO-OOH
DTPA
EDTA
ELISA
ESR

GSH

Hepes

hfs
MDA
MFO

NADPH

PBS
PUFA
SDS
SOD

TBA

ABBREVIATIONS

adenosine-5'-diphosphate
bicinchoninic acid
diethyldithiocarbamate
5.5'-dimethyl-1-pyrroline-N-oxide
DMPO spin trap adduct with -0H
DMPO spin trap adduct with 02?
diethylenetriamine penta-acetic acid
ethylenediaminetetracetate

enzyme linked immunosorbent assay
electron spin resonance
glutathione, reduced form

A-(Z-hydroxyethyl)-1-piperazine
ethanesulfonic acid

hyperfine splitting
malondialdehyde
mixed function oxygenase

reduced nicotinamide adenine
dinucleotide phosphate

phosphate buffered saline
polyunsaturated fatty acid
sodium dodecyl sulfate
superoxide dismutase

2-thiobarbituric acid

xi

INTRODUCTION

A free radical can be defined simply as any species possessing an
unpaired electron. The need to pair electrons imparts upon free
radicals their characteristic reactivity towards a variety of organic
and inorganic molecules. Since the demonstration by McCord and
Fridovich in 1969 that radical Species of molecular oxygen exist in
biological systems, a plethora of literature reports on the generatbon
and consequences of oxygen radicals has appeared.

This topic is of great interest as it indicates that survival in
an oxygen environment presents a perplexing dilemma to aerobic
organisms. While molecular oxygen is required for normal metabolic
function, the potential for cells to produce reactive, partially
reduced species of dioxygen necessitates careful control over the
generation of such species. Consequently, cells have evolved a series
of concerted mechanisms which control the production of, or limit the
accumulation of oxygen radicals. For example, electron transfer
pathways such as that of the mitochondria appear to be tightly coupled
such that the ultimate product of dioxygen reduction is water, with
little escape from the organelle of any of the partially reduced

oxygen species. In addition, all oxygen utilizing organisms possess

enzyme systems which serve to remove or "detoxify" potentially
deleterious oxygen radicals.

Perhaps one of the most important means by which dioxygen
activation can be controlled is by preventing its interaction with
transition metals such as iron.. Dioxygen exists as a diradical in the
ground state, containing two unpaired electrons of parallel spin in
the 2px” antibonding orbital. Fortunately, it is this property which
limits its direct reaction with organic molecules as this would
require compounds to possess two electrons of parallel spin, in
contradiction to the Pauli Exclusion Principle. However, transition
metals contain unpaired electrons and can readily form complexes with
dioxygen resulting in the production of partially reduced oxygen
species. Therefore, one would expect that a critical control point
for limiting the generation of active oxygen would be to prevent
interactions between oxygen and transition metals. Accordingly, the
transport and storage of these metals by Specialized proteins is a
highly regulated process which serves to minimize inadvertent oxygen
reduction.

While cellular production of oxygen radicals has been
intensively studied, there is an increasing awareness of a lack of
knowledge concerning the potential for biological iron-containing
proteins to participate in oxygen radical reactions. However, it is
becoming more apparent that iron-oxygen interactions do occur 13 3119.
In particular, iron-dependent peroxidation of membrane polyunsaturated
fatty acids is implicated in a variety of toxicological and
pathological states. In this dissertation, evidence is presented that

ferritin, the major iron storage protein, can provide iron for the

formation of radical species capable of initiating membrane lipid
peroxidation. Chapter I describes studies which demonstrate that 027.
generated by xanthine oxidase, can reductively release iron from
ferritin. Once released from ferritin this iron can promote the
peroxidation of phospholipid liposomes. ESR spin trapping studies
demonstrate an inverse correlation between -OH formation and lipid
peroxidation.

In Chapter II the cyclic reduction and autoxidation of paraquat,
catalyzed by NADPH-cytochrome PASO reductase. is shown to generate 027
and release iron from ferritin. The inability of fOH scavengers to
inhibit lipid peroxidation again provides evidence that lipid
peroxidation may be initiated by a Fe2+-02-Fe3+ complex, rather than
-OH.

A study of iron release from ferritin by paraquat (Chapter III)
revealed that the reduced paraquat cation radical can mediate a very
rapid, complete release of iron from ferritin. ESR and visible
Spectrophotometric studies demonstrated a rapid transfer of electrons
from the radical species of paraquat to ferritin, thereby reducing and
releasing the iron. The ferritin protein structure was not altered by
the radical and retained its ability to take up additional iron. The
semiquinone free radicals of the cardiotoxic anthracycline
antibiotics, adriamycin and daunomycin, are also capable of
reductively releasing iron from ferritin (Chapter IV). Importantly, a
relatively non-toxic autitumor drug, diaziquone, is also reduced to a
semiquinone free radical but is unable to mediate iron release from

ferritin.

The last Chapter (V) summarizes results obtained from a careful
study of the role of endogenous iron in microsomal lipid peroxidation.
Nearly 90% of microsomal iron was found to be associated with ferritin.
A procedure for ferritin removal was devised and the ability of
ferritin to promote microsomal lipid peroxidation was investigated.

Each of the aforementioned chapters is written in a format
similar to that of many scientific journals, thus each chapter
contains an Abstract, Introduction, Materials and Methods, and
Discussion as well as its own List of References. As the subject
matter of this dissertation encumbers a variety of topics including
oxygen radicals, lipid peroxidation, and iron metabolism a Literature
Review detailing each of these topics, and containing its own List of

References, precedes Chapter I.

LITERATURE REVIEW

Univalent Reduction of Molecular Oxygen

1. Superoxide and Hydrogen Peroxide. The sequential one
electron reduction of molecular oxygen produces superoxide (027),

hydrogen peroxide (H202), hydroxyl radical (fOH) and, ultimately,

water.
e- e- e- e-
024027+H2024.00H—¥>H20 (1)
2H H+ ‘

Superoxide is known to be produced by many enzyme systems such as
xanthine oxidase and flavin dehydrogenases. Many enzyme systems
known, or prOposed to generate 027 and H202 are dependent upon iron as
a means of reducing dioxygen. Xanthine oxidase, which oxidizes
xanthine and hypoxanthine to uric acid, contains iron sulfur centers
as well as molybdenum and FAD. This electron transport system is
capable of storing either 6 (1), or more recently, 8 electrons (2).
Its ability to produce both 027 and H202 has led to its widespread use
in oxygen radical-related research. Stimulated polymorphonuclear
leukocytes also produce large fluxes of 027 as a desired product of
their electron transport system, which utilizes a b type cytochrome
(3). This generation of 027 is thought to be required for the
bactericidal preperties exhibited by these leukocytes.

5

The inadvertent reduction of dioxygen by transition
metal-dependent electron transport systems, evolved for other
metabolic functions in the cell, has been prOposed. The iron and
copper containing terminal oxidase of the mitochondrial electron
transport chain, cytochrome c oxidase, catalyzes the A electron
reduction of molecular oxygen to water. However, H202 production by
mitochondria has been demonstrated and is thought to arise from
autoxidation of reduced intermediates of the system, particularly
ubisemiquinone (A,5). Likewise, generation of 027 and H202 by the MFG)
system of the endoplasmic reticulum has also been reported (6,7).

This electron transport system, which consists of two cytochromes and
their respective flav0proteins, functions to insert one atom of
molecular oxygen into the substrate (a host of endogenous compounds
and xenobiotics) with the other oxygen atom reduced to water. The
ability of inducers or uncouplers of MFO activity to enhance 027
and/or H202 production and conversely, of MFO inhibitors to decrease
their production, suggests that dioxygen reduction may result from
inefficient electron transfer (8). The significance of these findings
under normal cellular conditions remains to be determined.

The autoxidation of reduced, organic compounds such as
ubiquinone (A), epinephrine (9), and reduced flavins (10) has been
suggested to produce 027 and H202. Similarly, the simple autoxidation

of Fe2+ will also produce 027 (11):

Fe2+ + 02 -———>Fe3* + 023' (2)

As the reaction of dioxygen with organic molecules is kinetically
unfavorable, the proposed autoxidation of many compounds may also be
related to the presence of adventitous transition metal ions.

While simplistic in theory, Fe2+ autoxidation is subject to
variation, particularly in the presence of other ions or chelators.
The influence of chelation on Fe2+ autoxidation is usually manifested
as alterations in 1) the rate of autoxidation, 2) which reduced oxygen
species is produced, and 3) the redox potential of the metal. The
ability of numerous anions including chloride (12), sulfate (13), and
phosphate (1“) to affect Fe2+ autoxidation has generally indicated
that autoxidation increases as the affinity of the anion for Fe3+
increases. The complexity of these effects is demonstrated by the
ability of certain anoins to destabilize the Fe2*-02 complex resulting
in 027 formation while others stabilize the complex, allowing reaction
with a second Fe2+ to generate H202.

While H202 can be produced directly by Fe2+ autoxidation without
027 as an intermediate product, the majority of non-enzymatic H202
production likely involves formation of 027 initially. Once 027 is
generated, H202 can be formed via several pathways, for example, by
reaction with Fe2+z

211*
be?“ + 02:——>Fe3+ + H202 (3)
Hydrogen peroxide is also readily formed by non-enzymatic dismutation
of 027 or the action of SOD as shown in reaction (A):

211*
02: + 02:————->11202 + 02 (11)

At physiological pH the second order rate constant for the
non-enzymatic disproportionation of 027 is 8 x 10” M'1s"1 while the
SOD-catalyzed reaction proceeds at 2 x 109 M’1s'1 (15). Superoxide
dismutase actually refers to three related, yet distinct enzymes which
differ primarily with respect to the redox active metal utilized to
catalyze dismutation. The cytosolic SOD contains Cu2+ and Zn21, a
mitochondrial SOD contains Mn2*, while an Fe3+-containing variant has
been found in bacteria (15). These enzymes exemplify the use of
metals in controlled oxygen radical reactions and it is often proposed.
that their existence in nearly all aerobic organisms is testimony to
the importance of controlling the cellular flux of 027.

Hydrogen peroxide is also known to be generated by many enzymes
such as D-amino acid oxidase and urate oxidase which are localized
within peroxisomes. Not surprisingly, this organelle contains high
levels of catalase, a hemoprotein which serves to disproportionate
H202 to water and dioxygen. However, calculations show no: of the
generated H202 can diffuse out of the peroxisome (16). Many
peroxisomal proliferating or hyperlipidemic drugs are proposed to
result in the excessive production of H202 which may then "leak" out
of the peroxisome (17). While this process is thought to damage
cells, whether it actually occurs is questionable and furthermore,
cells also contain the cytosolic, selenium-containing enzyme

glutathione peroxidase which can remove intracellular H202 (18).

2. Formation of the Hydroxyl Radical. As the formation of H202

involves the addition and pairing of two electrons in the w”

antibonding orbital of dioxygen, H202 is not considered a radical

although it is a strong oxidizing agent and is generally included in
the category of "active oxygen". However, H202 readily reacts with
transition metals to generate oxidants capable of reacting with
numerous organic molecules. One of the most well known reactions is

that of H202 with Fe2+, known as the Fenton's reaction (19):
Fe” + H202——"Fe3+ + :011 + “on (5)

In fact, Fenton's reagent has been used for many years as a proficient
hydroxylating reagent as fOH is a very reactive species capable of
reacting with virtually all organic molecules in aqueous media, at
nearly diffusion controlled rates. Reaction types of -OH can be
hydrogen abstraction, addition, or electron transfer and its
indiscriminate reactivity has led to proposals of its reaction with
numerous biomolecules including phospholipids, proteins, and DNA (20).
Numerous 13 11339 studies have demonstrated that, as with
autoxidation, chelation can markedly affect the reaction of metals
with H202. Unchelated Fe2+ readily promotes {OH formation while
chelates which lack a free coordination site such as
DTPA and desferrioxamine apparently prevent fOH generation (21). On
the other hand, chelation of Fe2+ by EDTA may increase the yield of
-OH by 1) producing additional H202 by rapid autoxidation or 2)
maintaining the resulting Fe3+ in a soluble form (22). However, in the
presence of adequate amounts of H202, EDTA may also decrease fOH
production by facilitating the autoxidation of Fe2+, in which case

Fe2+ availability becomes rate limiting.

10

Lipid Peroxidation

 

1. An Overview of Initiation, Propagation and Termination. A
dramatic example of the concerted interaction of dioxygen and
transition metals in promoting oxidative stress is the peroxidation of
PUFA of membrane lipids. The lowered bond dissociation energy of the
allylic hydrogens of their methylene carbons renders PUFA more
susceptible to oxidative damage. The overall process of lipid
peroxidation can be divided into three main phases: initiation,
propagation, and termination (23). Initiation refers to the
abstraction of an allylic hydrogen from the PUFA of the phospholipid
(LH) by an oxidant, perhaps one of the partially reduced species of
oxygen. This results in the production of a carbon centered lipid
radical (Lf):

LH———->L- (6)
-H.
Subsequently, L- undergoes rearrangement (diene conjugation) and
reacts rapidly with molecular oxygen to produce a lipid peroxyl

radical (LOOf):
L- + 02———>L00- (7)

The propagative phase of lipid peroxidation begins when L00-
reacts with another divinyl methane of the same molecule or of a
neighboring PUFA, generating a lipid hydroperoxide (LOOH) and another

Lf:

1.00: + LH-———>LOOH + L-_ (8)

11

Propagation of lipid peroxidation is also promoted by transition metal
ions which can cleave LOOH to reactive L00- or alkoxy lipid radicals

(L0-) which can also react with LH:

Mn+ + LOOH——->LO- + on“ + MIN) (9)

MUM” + LOOH———>LOO- + 11* + MW (10)

Certain hemoproteins may also promote the peroxidative process by
cleaving LOOH (2A). The generation of reactive intermediates during
the propagative phase of lipid peroxidation, which themselves can
abstract susceptible hydrogens, imparts an auto-catalytic nature to
the process. It has been estimated that 8 to 1A prOpagation cycles
occur for each free radical generated (25). Hydroperoxides are often
formed during isolation and preparation of phospholipids and, in the
presence of transition metals, these PUFA readily undergo
LOOH-dependent peroxidation. It is likely that discrepancies among
laboratories with regard to initiation of lipid peroxidation via
hydrogen abstraction is due to contaminating LOOH.

Lipid peroxyl radicals are also known to undergo internal
cyclization to form a PUFA endoperoxide shifting the unpaired electron
to a carbon center. On to this, dioxygen again rapidly adds to form a
PUFA endoperoxide peroxyl radical which can then propagate the
peroxidative process in the same manner as LOOf. Additionally, this
radical can undergo a series of cleavage reactions to yield a diverse
array of products including aldehydes, ketones, alcohols, and ethers.
One of the predominant products from PUFA containing at least 3 double

bonds is MDA (26). Malondialdehyde forms a Schiff base product with

12

TBA (1 MDA: 2 TBA) which absorbs maximally at 532 nm (27). While not
totally specific for MDA, this assay is the most widely used index of
lfl.ll££2 lipid peroxidation.

Termination reactions or defense mechanisms are either primary or
secondary. Primary defense mechanisms are those that prevent the
formation of initiators of lipid peroxidation. These can involve
limiting the generation or accumulation of 027 and H202 by 1)
efficient electron transport systems, or 2) the action of SOD,
catalase, and glutathione peroxidase. Alternatively, the binding of
iron to proteins such as transferrin and ferritin, to minimize its
potential reaction with dioxygen and its reduction products, may
represent an important defense mechanism (28).

Secondary mechanisms for controlling the peroxidative process
exert their effect on the propagative phase. One of the most
important defense systems is vitamin E (a-tocopherol) which is
strongly lipOphilic and integrates intimately with membrane
phospholipids. Its phenolic group donates hydrogen atoms to L: or
LOOf, thereby interrupting the free radical cascade of lipid
peroxidation (29). The resultant tocopherol semiquinone chromanoxyl
radical is fairly stable and may be reduced by ascorbate (30) or GSH
(31), regenerating a-tocopherol. Also considered a secondary defense
is glutathione peroxidase, which by reducing LOOH to the less
reactive, corresponding alcohol minimizes the probability of further
propagation.

The importance of initiation in lipid peroxidation is highlighted
by the tremendous amount of research effort directed towards

identifying oxidants capable of mediating methylene hydrogen

13

abstraction. Excessive intracellular generation of 027 and H202 by
numerous toxic chemicals capable of undergoing cyclic reduction and
autoxidation has been proposed to result in lipid peroxidation (32).
However, it is known that neither 027 nor H202, in aqueous media at
physiologic pH, is of sufficient reactivity to initiate the
peroxidative process (33). The undissociated perhydroxyl radical
(H021), with a pKa of ”.8, is capable of initiating peroxidation of
hydrOperoxide-free PUFA but only 0.25% of any 027 will be present in
this form at pH 7.A (3“). It may be important in the acidic,
phagocytic vacuole or in the vicinity of the hydrophobic membrane
matrix.where the pH is appreciably lower.

The limited reactivity of 027 and H202 suggests that lipid
peroxidation requires the formation of another, more potent oxidant
within cells. At present, there are at least two_widely prOposed
mechanisms for the formation of an initiator of lipid peroxidation,
both of which are highly dependent upon the redox chemistry of
transition metal ions. The two most intensively researched metals
have been iron and copper, which may be expected considering their
biological abundance. Some of the first evidence for the critical
involvement of iron in lipid peroxidation arose from the fortuitous
discovery by Hochstein gt 2l° (35) that MDA formation occurred in rat
liver microsomes during NADPH oxidation. It was subsequently
determined that this process was attributable to endogenous iron
present in the ADP being used (36). Rates of microsomal lipid
peroxidation were subsequently shown to be markedly affected by
washing microsomes (37,38) or by the addition of iron chelators to the
system (39). The source, or the nature of, this endogenous iron which

is found associated with microsomes is yet to be determined, however.

14

2. Initiation by the Hydroxyl Radical. Initiation of lipid
peroxidation by fOH has received considerable support. Its generation
is most often proposed to occur via the iron-catalyzed Haber-Weiss

reaction (19,MO):

 

 

027 + i-‘e3+ ' {Fey + 02 (11)
211+

02': + 027 47 H202 ‘0' 02 (11)

Fe2+ + H202 -——> i-"e3+ + 1011 + “0H (5)

Unfortunately, fOH involvement is generally invoked as a result of the
ability of SOD, catalase, and fOH scavengers such as mannitol and
benzoate to inhibit peroxidation. In this manner xanthine oxidase
dependent peroxidation of arachidonic acid was proposed to be
initiated by -OH (N1) as peroxidation was inhibited by SOD (inhibits
reaction 11), catalase (inhibits reaction 5), and mannitol.

Similarly, inhibition of microsomal NADPH-dependent lipid peroxidation
by these agents was taken as evidence for fOH-dependent initiation
(A2).

Fenton's reagent (reaction 5) was capable of initiating
peroxidation when the iron was chelated with EDTA (22). Conversely,
unchelated Fe2+ appeared to initiate peroxidation solely by cleavage
of pre-existing LOOH (H3) although other work (AA) suggests that at
lower, more physiologically relevant levels of iron, fOH formation
occurs readily with free Fe21. Continuing studies (A5) have provided

evidence that at a strict EDTA:Fe2+ ratio of one, 10H formation is

favored and accordingly, lipid peroxidation is observed. At EDTAzFe2+

ratios greater than one, lipid peroxidation is inhibited while at

15

ratios of less than one, EDTA-Fe2+ functions to decompose LOOH,
generating reactive Lo- which are then capable of rapidly peroxidizing
other PUFA. Similar results were obtained by Tien gt al. (A6)
however, they suggested that the optimal rates of lipid peroxidation,
noted at a ratio of less than one, were due to optimum fOH generation
resulting from a rapidly autoxidizing pool (EDTA-Fe2+) and the slower
oxidizing free Fe2+.

The aforementioned studies are indicative of the many factors
which can affect fOH production, even in what appears to be a
relatively simple system. They also demonstrate that several
interpretations of similar data are possible. In spite of these
shortcomings, the purported formation of fOH in a variety of even more
complex systems, subject to many more variables, is touted as indirect
evidence for -OH-dependent initiation of lipid peroxidation. McCord
and Day (A7) have demonstrated, using a xanthine oxidase system to
generate 027, that EDTA-Fe3+ effectively catalyzes fOH formation which
has since been confirmed by others (A3,Hfl,u8). Hydroxyl radical
formation with di- and triphosphate nucleotides, in the presence of
Fe2*, was demonstrated by Floyd (”9.50) with ESR spin trapping. This
was postulated to be a function of decreased autoxidation of chelated
Fe2+, thereby facilitating interaction with H202. Even more
speculative is the report which suggests that HZOZ-dependent fOH
formation, catalyzed by trace amounts of cellular iron, was

responsible for killing of mouse 3T3 cells (51).

3. Alternatives to the Hydroxyl Radical in Initiation. While

 

support for the involvement of fOH in lipid peroxidation continues,

16

there is a growing body of evidence which suggests that other
oxidants, particulary an iron-bound oxygen complex, may in many
instances be the actual initiator. Some of the first evidence
refuting jOH involvement was provided by Pederson and Aust (52) who
demonstrated that NADPH-dependent microsomal lipid peroxidation was
not inhibited by SOD or catalase and similarly, Hochstein and Ernster
(35) found catalase to be without effect. It is known that microsomes
contain endogenous catalase and significant accumulation of the H202
required to promote the Haber-Weiss reaction is doubtful. The
addition of exogenous catalase or azide, a catalase inhibitor, to
microsomes also had no effect on the rate of lipid peroxidation (53).
Recently, direct quantitation of 027 generation by microsomes has
shown it to account for only a minor percentage of their total
reduction potential, suggesting that fOH formation may not adequately
account for the initiation of microsomal lipid peroxidation (5A).

In lieu of the above findings it has been proposed that
initiation may be mediated by an oxidant other than fOH. In a
reconstituted peroxidation system employing liposomes prepared from
extracted microsomal lipid, and xanthine oxidase to generate 02?,
nucleotide-chelated Fe3+ was capable of initiating peroxidation (55).
Inhibition of peroxidation by SOD, but no effect of catalase or
mannitol, suggests initiation to be via a chelated, reduced
iron-oxygen complex such as the perferryl ion (or ferrous dioxygen,
FeZ+-02) originally proposed by Hochstein and Ernster (35).

The identification of NADPH-cytochrome PASO reductase as the
microsomal enzyme catalyzing the transfer of reducing equivalents to

Fe3+ chelates, necessary for lipid peroxidation, made it possible to

17

study NADPH-dependent lipid peroxidation in a liposomal system (56).
In contrast to the xanthine oxidase system, the addition of ADP-Fe3+
resulted in no peroxidation. Activity was dependent upon the
inclusion of EDTA-Fe3+ to this system. This dependence upon EDTA-Fe3+
is due to an inability of NADPH-cytochrome PASO reductase to generate
significant amounts of 027, or to directly reduce ADP-Fe3+ (5A).
However, the enzyme is capable of directly reducing EDTA-Fe3+ and
electron transfer from EDTA-Fe2+ to ADP-Fe3+ results in the formation
of an ADP-Fe21-02 complex capable of initiating peroxidation.
Accordingly, SOD, catalase and mannitol had no effect on rates of MDA
formation (52).

In an effort to further elucidate the nature of this postulated
Fe2*-oxygen complex, peroxidation initiated directly by Fe2*-chelates
was evaluated in hopes that elimination of a reducing agent or system
would simplify the situation. As discussed, Fe2+ autoxidation can be
quite complex and markedly affected by the nature of the chelating
ligands, however, it is reasonable to assume that all autoxidations
will produce some of the partially reduced forms of oxygen or their
iron-bound equivalent forms. The ability of EDTA- or DTPA-chelated
Fe2+ to generate fOH and to decompose peroxides has been reported.
Accordingly, peroxidation initiated by these Fe2+-chelates was
sensitive to SOD or catalase (57,58). Ferrous chelates of possible
physiological significance, such as phosphate, ADP, oxalate, and
citrate, initiated peroxidation but were largely unaffected by SOD or
catalase, suggesting formation of an initiator other than fOH

Lipid peroxidation initiated by nucleotide-chelated Fe2+

consistently exhibited a significant, initial lag phase before MDA

18

formation was observed (59). The addition of chelated-Fe31, under
conditions that minimized Fe2+ autoxidation, abolished the lag period
in a concentration-dependent fashion. Recent work in our laboratory
has demonstrated similar results with other FeZ+-chelates, and that
the rate and extent of lipid peroxidation is maximal at a Fe3":Fe2+
ratio of one (60). Of great interest was the finding that mannitol
and benzoate, which are thought to inhibit lipid peroxidation by
scavenging fOH, significantly affected Fe2+ autoxidation and therefore
affected lipid peroxidation by altering the Fe3*:Fe2+ ratio. These
data suggest that the initiator of lipid peroxidation may not be a
Fe2*-02 complex, but rather a Fe2+--Oz-Fe3+ complex of undetermined
structure or reactivity.

The inability of -0H scavenging agents to inhibit peroxidation in
systems employing nucleotide~chelated iron does not exclude fOH
formation in the system but suggests initiation to be via another
species. Indeed, {OH formation by nucleotide-Fe2+ chelates has been
described. One of the arguments against fOH as an initiator of lipid
peroxidation capitalizes on its extreme reactivity. While clearly
capable of hydrogen abstraction, it is likely to react at, or close
to, its site of generation. For example, Tien gt El° (22) were unable
to demonstrate lipid peroxidation in a liposomal system in the
presence of EDTA—Fe3+ and xanthine oxidase. However, when the
phospholipid was dispersed with detergent or replaced with
detergent-dispersed linoleate, peroxidation occurred (A6). Catalase,
mannitol and SOD significantly inhibited activity implying the

involvement of fOH. The inability of this system to initiate

19

The propensity of °OH to initiate peroxidation only in highly
artificial systems (e.g. synthetic iron chelators and
detergent-dispersed lipid) leads one to question the role of To“.lfl
1132- As an alternative to fOH, several investigators have proposed
the existence of a similar species known as crypto {OH (61,62). Its
generation involves the homolytic cleavage of H202 in which the
radical species formed remains constricted within its region of
generation; such as in an enzyme crevice. Others envision the fOH to
be "surrounded" by solvent forming a cage-like structure. These type
of restrictions may impart a degree of selectivity towards potential
substrates such as the allylic hydrogens of PUFA. Indirect evidence
for damage by such complexes, or by a "site-directed" Fenton's
reaction, where the metal is bound to a biological macromolecule and
cleaves H202 in close proximity to susceptible moieties such as
phospholipids, has been presented (63,6A).

It is readily apparent from all of the above studies that iron is
intimately involved in the initiation and propagation of lipid
peroxidation. While the true identity of an initiator of lipid
peroxidation remains elusive, a critical role of iron is implicit in
virtually all proposals. Most 1g 21359 studies employ low molecular
weight iron chelates, particularly ADP and EDTA, and much of this work
has demonstrated the dramatic effects which chelation can impart to
the peroxidative process. This knowledge has led to an increasing
awareness of the need to isolate and identify physiologic iron
chelates, and to assess their potential for providing iron for

participation in the formation of an initiator of lipid peroxidation.

20

Iron Metabolism

 

1. Absorption and its Regulation. Although iron metabolism has
been well studied, the potential for known physiological iron sources
to provide iron for participation in redox activities is not well
known. Highly regulated cellular control of iron is necessitated not
only by its potential toxicity, but also by the very limited
solubility of free Fe3+ in aqueous solutions. These principles are
illustrated by the ability of cells to rapidly synthesize ferritin in
response to iron loading (65) and the elaborate siderophore-dependent,
iron storing capacity of many microorganisms (66). Extensive studies
examining iron metabolism indicate that the oxidation state of the
iron significantly affects its incorporation or release from
biological iron-containing molecules. Thus, the ability of organisms
to reduce cellular iron suggests the potential for initiation of lipid
peroxidation.

Mechanisms for the excretion of body iron are quite limited,
therefore regulation of iron absorption is an important control point
for maintaining acceptable iron levels. Accordingly, variations in
iron absorption appear to reflect changes in total body iron stores.
Means of controlling absorption are not fully understood but are
thought to be related to levels of ferritin in the mucosa or to the
total iron content of the mucosal cell (67).

Intestinal absorption of iron occurs predominately in the

proximal portion of the small intestine, particularly the duodenum

(68). Uptake of iron requires iron as FeZI. therefore reduction of

dietary iron is necessary (69). Recent evidence indicates that

21

xanthine oxidase in brush border cells promotes the incorporation of
iron into mucosal transferrin for tranSport to the portal blood stream
(70). Others have suggested that iron enters mucosal cells when
complexed with a low molecular weight membrane carrier (71-7”). In
most studies, however, it is accepted that a mucosal transferrin
ultimately transports dietary iron to the serosal cell surface.

2. Transport of Iron. Iron circulates in the blood bound to
transferrin, a single-chain glycoprotein capable of binding two Fe3+
(75). The uptake and release of iron from transferrin is not fully
understood, although several hypothesis have been proposed. Iron
binding by transferrin requires two bicarbonate anions which have been
shown to bind to the metal but their structural role in assembling the
metal site is unknown (76,77). The anion appears to bind weakly to
the protein prior to the iron, perhaps resulting in a conformational
change which facilitates iron binding. Interestingly, the binding
constant of apotransferrin for Fe2+ is 17 times less that of Fe3+
(78), again emphasizing the importance of the oxidation state of the
iron. Ceruloplasmin, which serves as the major transport protein of
copper has significant ferroxidase activity and has been suggested to
facilitate oxidation of Fe2+ for binding to transferrin (79).

Iron appears to enter cells with transferrin following receptor
mediated endocytosis (86,87). The iron is subsequently released from
transferrin in an acidic, endocytic vesicle with the apOprotein
recycled to the cell surface. The mechanism of iron release from the
ternary complex is unknown but the weak affinity of transferrin for
Fe2+ suggests that reduction of the complexed iron would facilitate

its release and most investigators also envision protonation of the

22

anion in the vesicle as a means of freeing bound iron. Studies with
various reductants suggest that iron release is dependent upon not
only the redox potential of the agent, but also upon its effect on
protein conformation (88). It is speculated that at low pH, decreased
stability of the Fe3*-transferrin complex makes the metal more
susceptible to reduction.

An alternative hypothesis proposes that cells contain low
molecular weight iron chelates which can remove the iron from the
protein and also function in its intracellular transport. It has been
demonstrated that phosphate containing compounds such as pyrophosphate
and ATP are effective in iron removal from transferrin (89) while in
rabbit reticulocytes rates of iron uptake from transferrin directly
correlate with ATP concentrations (90). Iron release from transferrin
which is dependent solely on chelation occurs quite slowly but is
enhanced significantly at lowered pH. Thus, irrespective of
mechanism, iron release from transferrin is likely to occur within the
acidic millieu of the non-lysosomal endocytic vesicle.

In spite of intensive study, mediators of intracellular iron
transport have not been positively identified although most
investigators suggest that a low molecular weight intermediate may
function in this respect (91). For example, phosphate compounds
involved in the removal of iron from transferrin may also shuttle iron
throughout the cell cytosol. It is this elusive "pool" that is
generally prOposed to be the most likely candidate for participating
in oxidative processes. While some evidence for its existence in
reticulocytes has been presented (92.93) definitive identification in
other cells is lacking although chelators such as ADP (42) and citrate

(94) have been proposed.

23

3. Iron Storage in Ferritin: Uptake and Release. Ultimately,
the vast majority of intracellular iron is stored in ferritin which is
predominately a cytosolic protein although very recent work indicates
an association with the endoplasmic reticulum. It is a large
(Mr-A50,000) protein comprised of 2A subunits arranged symmetrically
about a hollow central core (95). Iron is stored in this core with an
average compositon of (FeOOH)nge0f0PO3H2. Each ferritin molecule is
capable of binding up to A500 atoms of iron although ferritin is
normally only 20% loaded with iron (96). Access to the core is via 6
narrow hydrophobic channels and 8 smaller, hydrophilic channels which
average 0.9 nm to 1.2 nm in width (97).

The mechanisms of deposition and mobilization of iron from
ferritin remain enigmatic but appear to involve oxidation and
reduction, respectively (98). It is generally thought that deposition
of iron occurs after the protein subunits are assembled, because
apoferritin accelerates the rate of oxidation of Fez+ (99). The
electron acceptor £3 3119 is likely to be dioxygen although whether
02?, H202 or H20 is the product remains to be resolved (100,101).
Elucidation of mechanism(s) and identification of products is
complicated by evidence that oxidation occurs at two sites; one on the
protein and the other on the iron core, and that oxidation may occur
via different mechanisms at each site (102).

While it is accepted that mobilization of iron from ferritin
requires reduction, physiological reductant(s) remain unknown. The
”crystal growth theory" (103) proposes that reductants traverse the
length of the channels to directly reduce the ferric iron core while

others (10A) envision electron transfer along the channels, initiated

24

at oxido-reduction sites on the interior of the channels. Reduced
flavins, which are among the most effective mobilizers of iron from
ferritin, appear to at least partially enter the channels (105).
Therefore, size constraints may limit the number of potential
biological reductants.

Physiological reductants such as GSH, ascorbate and cysteine
release iron from ferritin at rates too slow to be considered of
significance (106) while non-physiological reducing agents such as
dithionite and thioglycolate readily release ferritin iron lg zitgg
(107.108). Certain chelating agents have been shown to facilitate
iron release from ferritin, but only in the presence of a reductant
(109). Enzymatic means for releasing iron from ferritin have been
suggested. Evidence has been presented for the existence of a
NADH-FMN oxido-reductase which provides reduced flavins to mediate
iron release but it is highly unstable and is not well characterized
(110,111). Several studies have also indicated that hepatic xanthine
oxidase, in its NAD-requiring dehydrogenase form, may play a role in
ferritin iron release (112,113). Another factor to be considered is
the redox potential of the reductant as recent work reports the
reduction potential for iron in ferritin to be approximately -230 mV
at pH 7.0 (11A). Accordingly, the efficacy of a series of reduced
flavins to mediate iron release from ferritin closely paralleled their
respective redox potentials (105).

A. Biological Iron and Lipid Peroxidation. The similarities in
the conditions under which the release of iron from ferritin and lipid
peroxidation are favored, i.e. the presence of reducing agent(s) and

chelator(s) is rather striking. Surprisingly, however, relatively few

25

studies have investigated the ability of ferritin to supply iron for
the promotion of lipid peroxidation. This is likely a result of the
general concensus that ferritin provides a secure means of storing
iron in an inert form. However, it is becoming increasingly evident
that ferritin may play a more dynamic role lg 2119. For example,
mitochondria contain distinct binding sites for ferritin from which
iron is mobilized (115,116). Gutteridge (117) reported that ferritin
promoted £2 21523 ascorbate-dependent lipid peroxidation while others
demonstrated fOH formation using ferritin and ascorbate (118).
Similarly, Wills (119) demonstrated that ferritin may promote
non-enzymatic peroxidation, however, it was subsequently suggested
(38) that ferritin was not responsible for the increase in lipid
peroxidation observed in liver microsomes isolated from iron-loaded
rats. Thus, in lieu of the rather cursory examination of the ability
of ferritin to promote lipid peroxidation, and the large amount of
intracellular iron stored within this protein, a more thorough
evaluation of this topic is warranted.

ESR spin trapping has demonstrated the ability of transferrin to
inhibit (80,81) or catalyze (82,83) fOH formation in the presence of a
027 generating system. This is related to the degree of iron
saturation as only fully loaded transferrin stimulates fOH production
and promotes lipid peroxidation. Partially loaded transferrin
inhibits both 70H generation and lipid peroxidation (8A,85).
Transferrin is normally only 301 saturated on the average, therefore
it may serve in a protective capacity.

In addition to ferritin and transferrin, much of the total body

iron is found in hemoglobin and myoglobin. While it is generally

26

concluded that these proteins would be unlikely to donate iron to
participate in oxygen radical formation, several recent studies have
addressed this topic. Hemoglobin can promote 027-dependent, jOH
formation and the peroxidation of red cell membranes (120). Other
studies have shown that methemoglobin and metmyoglobin can form
complexes with H202 which are apparently capable of initiating lipid
peroxidation (121)

Traces of iron are also present in extracellular fluids although
the nature of these complexes is unknown and their ability to catalyze
oxygen radical formation is controversial. Winterbourn (122) was
unable to detect jOH formation from plasma, lymph or synovial fluid
upon exposure to H202. In contrast, Gutteridge gt 2l° (123)
demonstrated the ability of synovial fluid to catalyze jOH formation
and promote lipid peroxidation. The physiological significance of
these findings is unknown, however, it has been shown that in certain
disease states, such as rheumatoid arthritis, the amounts of these
iron salts are elevated (12A,125). Correspondingly, a correlation
between iron levels and the formation of TBA-reactive material has

been demonstrated in the synovial fluid of these patients (126).
Iron in Proposed Oxygen Radical-Dependent Toxicities.

1. Paraquat. The toxicities associated with numerous drugs and
chemicals is often attributed to their ability to be metabolized to
free radical species, resulting ultimately in oxidative damage to
critical cellular constituents (32). For many structurally diverse
chemicals their metabolism is a virtue of their ability to serve as

alternate electron acceptors for the microsomal enzyme,

27

NADPH-cytochrome PASO reductase. The addition of a single electron to
these compounds produces an unstable, free radical intermediate that,
in most instances, reacts rapidly with dioxygen to produce 027 and
regenerate the parent compound, a process referred to as redox cycling
(32). As discussed, it is unlikely that the simple generation of 027
and H202 will initiate oxidative damage to membrane PUFA, proteins or
DNA. Thus, the toxicity of these chemicals is likely to require, or
be greatly potentiated by, the presence of transition metal ions.
Paraquat (methyl viologen, 1-1-dimethyl-A,A‘-bipyridylium
dichloride) is a widely used herbicide which is highly toxic to
mammals. Its herbicidal preperties are attributed to the reduction
within plant chlorOplasts of its bipyridyl, dication structure to a
mono-cation radical that readily autoxidizes to produce 027. Its
mammalian toxicity is presumably a result of reduction by
NADPH-cytochrome PASO reductase and subsequent redox cycling to

generate 027 (127):

92 curuQ—G- -013 NADPH
~ _- Cytochrome
Palm; ‘ P .450 Reductase

._. 4'
O2 013-": _0>—-<’__:N; ~01: NADP‘

28

Numerous investigations have attempted to correlate paraquat
toxicity with enhanced lipid peroxidation. While results are varied
it does appear that, in general, iron is involved in the toxicity of
paraquat. Many of the discrepancies between laboratories appears to
concern the type and amount of iron present in the various 12.11222
systems. Trush gt El' (128) demonstrated that paraquat stimulated rat
lung microsomal peroxidation in the absence of added exogenous iron.
However, peroxidation was completely inhibited by EDTA, thus trace
amounts of endogenous iron probably were responsible for promoting the
peroxidative activity. In a reconstituted liposomal peroxidation
system utilizing isolated NADPH-cytochrome PA50 reductase, paraquat
also stimulated peroxidation when supplied with ADP-Fe3+ (127).

Others studies have not supported lipid peroxidation as a
mechanism of toxicity for paraquat. One investigation reported no
effect of paraquat on iron-catalyzed microsomal lipid peroxidation
(129). Conversely, it has been suggested that paraquat inhibits lung
microsomal lipid peroxidation dependent upon endogenous iron by
interrupting microsomal electron transport which normally produces a
reduced iron-oxygen intermediate capable of initiating peroxidation
(130). These conflicts have not been totally resolved but may reflect
NADPH depletion by continuous redox cycling of paraquat in the studies
which report an inhibition of peroxidation. In fact, reduction of
NADPH levels 12.1izg has also been suggested as a mechanism of
toxicity for paraquat (131).

Recently, several 13 1119 studies have provided evidence that
paraquat toxicity is mediated, in part, by redox active metals. In

paraquat-treated mice, the co-administration of Fezsou greatly

29

potentiated toxicity (132). On the other hand, treatment of the
animals with the iron chelator desferrioxamine partially ameliorated
toxicity (132). With §;.§211v cell inactivation by paraquat was
significantly increased by small amounts of copper ions which enhanced
degradation of the cytoplasmic membrane (133).

If it is assumed that paraquat toxicity does occur via iron
catalyzed lipid peroxidation, the topic still to be addressed concerns
the nature of the initiating species. It has been demonstrated with
paraquat that, 12 [$229, trace amounts of EDTA- and DTPA-Fe3+
catalyzed fOH formation of which 971 was free to diffuse into solution
(13A). Kohen and Chevion have proposed "site-specific" fOH generation
to rationalize paraquat-dependent E; 9911 cell killing (133).
Treatment of paraquat intoxicated animals with fOH scavengers such as
dimethylthiourea has generally not proven beneficial (135). Very
recent work with Chinese hamster ovary cells reported that SOD
provided protection against paraquat damage but catalase was without
effect (136). Many other studies have provided very conflicting
evidence on the ability of SOD to prevent or lessen paraquat-induced
toxicity (137,138).

It can be seen that the only concensus that can be derived from
the literature is that paraquat toxicity, if it indeed does involve
lipid peroxidation, must require a source of iron. Thus, if lipid
peroxidation is to remain a viable alternative to account for
paraquat-dependent toxicity, it is imperative to identify
physiological sources of iron which may be released or made available
by paraquat, allowing the formation of an initiator of lipid

peroxidation.

3O

2. Adriamycin. There are a number of other toxicities in which

 

iron is thought to play a vital role. One of the most important, and
intensively studied, is the toxicity of adriamycin. This
anthracycline antibiotic is widely used as a chemotherapeutic agent.
However, its clinical use is limited by a dose-dependent
cardiomyopathy leading to congestive heart failure. There are several
proposed mechanisms for this selective cardiomyopathy, of which lipid
peroxidation appears to be currently favored. The quinone moiety of
adriamycin can be reduced by one electron to a semiquinone free
radical by several flavoproteins including NADPH-cytochrome PASO

reductase (139):

N A D PH
Cytochrome P450 Reductase
e- .

     

Subsequent redox cycling to generate 027 and H202 is thought to lead
to oxidative stress and eventually, cell death. The heart appears to
be particularly susceptible to oxidative damage due to low levels of

SOD and catalase relative to other oxygenated tissues (1A0).

31

Microsomal lipid peroxidation induced by adriamycin can be
inhibited by SOD, catalase, fOH scavengers, and iron chelators,
indicative of initiation via the iron-catalyzed Haber-Weiss reaction
(1A1). Recently it was shown that the concentration of ferrous ions
markedly affects adriamycin-induced peroxidation (1A2) while others
have suggested that in the absence of iron, adriamycin does not
potentiate microsomal lipid peroxidation (1A3). Evidence for similar
processes occurring 12.1112 was provided by Mimnaugh 23 El' (1AA) who
demonstrated that lg 1119 administration of adriamycin resulted in
biochemical changes which were manifested as an 18-fold increase in
NADPH-dependent cardiac microsomal lipid peroxidation. Also
implicating iron in anthracycline toxicity was the demonstration that
co-administration of an EDTA derivative provided some protection
against adriamycin-induced cardiac damage (1A5).

While a requirement for iron in adriamycin toxicity is generally
accepted, again the mechanism(s) by which an initiating species is
formed remains controversial. Elegant work by Nakano and co-workers
(1A6) has demonstrated that while adriamycin may potentiate lipid
peroxidation, the process is quite complex and could involve an
oxidant other than 70H. An absolute requirement of iron for
adriamycin-stimulated peroxidation was originally demonstrated by
these authors (1A6) who subsequently reported that adriamycin formed a
complex with ferric iron and ADP (1A7). This complex was capable of
undergoing "self-reduction" with the reduced complex capable of
initiating peroxidation (1A8) in a -OH-independent fashion. The
necessity for reduction of the iron in this complex is demonstrated by

the ability of the ferroxidase activity of ceruloplasmin to inhibit

32

complex-initiated peroxidation (1A9). Adriamycin is also capable of
mediating oxidative damage to DNA. No concensus has been reached with
regard to the nature of the damaging species with evidence presented
for both fOH (150) and an iron-oxygen complex (151).

In spite of these intensive research efforts which implicate iron
as a causative factor in adriamycin-induced cardiomyOpathy, the
develOpment of effective preventative or therapeutic regimens, or new
drugs, is limited by a lack of knowledge concerning the $2 3119 source
of iron which is accessible to adriamycin. Recent work by Demant
(152) has shown that the chelative properties of adriamycin allow it
to slowly mobilize iron from ferritin. Similarly, iron released from
transferrin at acidic pH could be chelated by adriamycin (153).
However, in both instances rates of iron accumulation by adriamycin
were very slow. These mechanisms also preclude reduction of the
anthracycline to its semiquinone free radical, which is regarded as a
prerequisite for its toxicity. Thus, a definitive correlation or
interaction between the redox cycling of adriamycin and biological

iron stores has yet to be established.

LIST OF REFERENCES

7.

10.

LIST OF REFERENCES

J. 8. Olson, D. P. Ballou, G. Palmer and V. Massey. The
mechanism of action of xanthine oxidase.‘ J. Biol. Chem. 2A9:
”363‘A382 (197A) ' '

R. Hille and V. Massey.‘ The presence of a reducible disulfide
bond in milk xanthine oxidase. J. Biol. Chem. 257: 8898-8901
(1982). ' ' ' '

D.R. Light, C. Walsh, M.O. O'Callaghan, E.J. Goetzl, and A.I.
Tucker. Characteristics of the cofactor requirements for the
superoxide-generating NADPH oxidase of human polymorphonuclear
leukocytes. Biochem. 20: 1A68-1A76 (1981).

A. Boveris, E. Cadenas, and A.0.M. Stoppani. Role of
ubiquinone in the mitochondrial generation of hydrogen
peroxide. Biochem. 156: ABS-AAA (1976).

E. Cadenas, A. Boveris, I. Ragan and A.0.M. Stoppani.
Production of superoxide radicals and hydrogen peroxide by
NADH-ubiquinone reductase and ubiquinol-cytochrome 0 reductase
from beef heart mitochondria. Arch. Biochem. Biophys. 189:
2A8-257 (1977) ' - -

H. Kuthan, H. Tsuji, H. Graf, V. Ullrich, J. Werringloer and
R.W. Estabrook. Generation of superoxide as a source of
hydrogen peroxide in a reconstituted monooxygenase system.
FEBS Lett. 21: 3A3'3A5 (1978).

J. Werringloer. The formation of hydrogen peroxide during
hepatic microsomal electron transport reactions. In:
Microsomes and Drug Oxidations (V. Ullrich, I. Roots, A. G.
Hildebrandt, R. W. Estabrook, and A. H. Conney, eds. .)' pp.
261-268, Pergamon Press, Oxford (1977).

H.G. Thurman, H.G. Ley and R. Scholz. Hepatic microsomal
ethanol oxidations: Hydrogen peroxide formation and the role
of catalase. Eur. J. Biochem. 25: A20-A30 (1972).

H.P. Misra and I. Fridovich. The role of superoxide anion in
the'autoxidation of epinephrine and a simple assay for
superoxide dismutase. J. Biol. Chem. 2A7: 3170-3180 (1972).

V. Massey, G. Palmer and D. Ballou. 0n the reaction of
reduced flavins and flav0proteins with molecular oxygen. In:
Flavins and Flavoproteins (H. Kamin, ed. ), pp. 3A9-361,

 

University Park Press, Baltimore, MD (1971).

33

11.

12.

13.

1A.

15.
16.

17.

18.

19.

20.

21.

22.

23.

34

G. Just. Kinetische untersuchung der autoxydation des in
wasser gelosten ferrobicardonats. Che. Ber. Ag: 3695-3701
(1907). ‘ ‘

A. M. Posner. The kinetics of autoxidation of ferrous ions in
concentrated HCl solutions. Trans. Faraday Soc. 32: 382-388
(1953). ' '

A.B. Lamb and L.W. Elder. The electromotive activation of
oxygen. J. Am. Chem. Soc. 23: 137-163 (1931).

M. Cher and N. Davidson. The kinetics of the oxygenation of
ferrous iron in phosphoric acid solution. J. Am. Chem. Soc.
77: 793-798 (195AL ‘ ‘ '

I. Fridovich. Superoxide radical: An endogenous toxicant.
Ann. Rev. Pharmacol. Toxicol. 23: 239-257 (1983). '

A. Boveris, N. Oshino and B. Chance. The cellular production
of hydrogen peroxide. Biochem. J. 128: 617-630 (1972).

J.K. Reddy, J.R. Warren, M.K. Reddy and N.D. Lalwani. Hepatic
and renal effects of peroxisome proliferators: Biological
implications. Ann. N. Y. Acad. Sci. 386: 81-110 (1982).

R.A. Condell and A.L. Tappel. Evidence for the suitability of
glutathione peroxidase as a protective enzyme: Studies of
oxidative damage, renaturation, and proteolysis. Arch.
Biochem. Biophys. 223: A07- A16 (1983L ‘

F. Haber and J. J. Weiss. The catalytic decomposition of
hydrogen peroxide by iron salts. Proc. Roy. Soc. London Ser.
A. 1A7: 332-351 (193A). - ' '

R.L. Willson. Free radicals and tissue damage: Mechanistic
evidence from radiation studies. In: Biochemical Mechanisms
of Liver Injury (T. F. Slater, ed. ),p pp. 123-22A, Academic

 

 

Press, New York (1978).

E. Graf, J.R. Mahoney, H.G. Bryant and J.W. Eaton.
Iron-catalyzed hydroxyl radical formation:' Stringent
requirement for free iron coordination site. J. Biol. Chem.
259: 3620-362A (198A). ‘ '

M. Tien, B.A. Svingen and 8.0. Aust. An investigation into
the role of hydroxyl radical in xanthine oxidase-dependent
lipid peroxidation. Arch. Biochem. BiOphys. 216: 1A2-151
(1982). ‘ ‘_' ‘ -

J.F. Mead. Free radical mechanisms of lipid damage and
consequences for cellular membranes. In: Free Radicals in
Biology, Volume I (W. A. Pryor, ed. ), pp. 51-68, Academic
Press, New York (1976).

 

 

2H.

25.

26.

27.

28.

29.

30.

31.

32.

33.

3”.

35.

36.

35

P.J. O'Brien. Intracellular mechanisms for the decomposition
of a lipid peroxide. I. Decomposition of a lipid peroxide by
metal ions, heme compounds, and nucleophiles. Canadian J.
Biochemistry #7: “85-492 (1969). ‘

G.S. Wu, R.A. Stein and J.F. Mead. Autoxidation of fatty acid
mono-layers absorbed on silica gel: II. Rates and products.
Lipids 12: 971-978 (1978). '

N. A. Porter, J. Nixon and R. Isaac. Cyclic peroxides and the
thiobarbituric assay. Biochim. BiOphys. Acta uu1: 506-512
(1976). - ' “‘~ -

J.A. Buege and S.D. Aust. Microsomal lipid peroxidation. In:
Methods in Enzymology, Volume 52 (S. Fleischer and L. Packer,
eds.). DP. 302-310, Academic Press, New York (1978).

R.R. Crichton. Interactions between iron metabolism and
oxygen activation. In: Oxygen Free Radicals and Tissue

Damage, pp. 57-76, Excerpta Medica, Amsterdam (1979).

A.L. Tappel. Vitamin E and free radical peroxidation of
lipids. Annal. New York Acad. Sci. 203: 12-28 (1972).

E. Niki, T. Saito, A. Kawakami and Y. Kamiya. Inhibition of
oxidation of methyl linoleate in solution by vitamin E and
vitamin C. J. Biol. Chem. 259: ”177-”182 (198“).

C.C. Reddy, R.W. Soholz, C.E. Thomas and E.J. Massaro.
Vitamin E dependent reduced glutathione inhibition of rat
liver microsomal lipid peroxidation. Life Sci. 31: 571-576
(1982). ‘ ‘ '

H. Kappus and H. Sies. Toxic drug effects associated with
oxygen metabolism. Redox cycling and lipid peroxidation.
Experentia 31: 1233- 12M1 (1981).

J.A. Fee and J.S. Valentine. Chemical and physical properties
of superoxide. In: Superoxide and Superoxide Dismutases (A.M.
Michelson, J. M. McCord and I. Fridovich, eds. ), pp. 19-60,
Academic Press, New York (1977).

 

B. H. J. Bielski, R. L. Arudi and M. W. Sutherland. A study of
the reactivity of Hog/02 with unsaturated fatty acids. J.
3101. Chem. 258: H759- 1761 (1983).

P. Hochstein and L. Ernster. ADP-activated lipid peroxidation
coupled to the TPNH oxidase system of microsomes. Biochem.
Biophys. Res. Comm. 12: 388-394 (1963). '

P. Hochstein, K. Nordenbrand and L. Ernster. Evidence for the
involvement of iron in the ADP-activated peroxidation of
lipids in microsomes and membranes. Biochem. BiOphys.

Res. Comm. 15; 323'328 (196“). ‘

37.

38.

39.

“0.

N1.

H2.

"3.

H".

”5.

N6.

47.

H8.

36

J.L. Poyer and P.B. McCay. Reduced triphosphOpyridine
nucleotide oxidaseécatalyzed alterations of membrane
phospholipids. J. Biol. Chem. 2N6: 263-269 (1971).

S.D. Hills. Lipid peroxide formation in microsomes. The role
of non-haem iron. Biochem. J. 113: 325-332 (1969).'

E.D. Wills. Mechanisms of lipid peroxide formation in tissues.
Role of metals and hematin proteins in the catalysis of the
oxidation of unsaturated fatty acids. Biochem. BiOphys. Acta
98. 238- 251 (1965) ' '

J.A. Fee. On the question of superoxide toxicity and
the biological function of superoxide dismutases. In:
Oxidases and Related Redox Systems (T.E. King, H.S. Mason and

 

M. Morrison, eds. ),p pp. 101-1H9. Pergamon Press, New York
(1982) ' ' ‘ ' '

S. E. Fridovich and N. A. Porter. Oxidation of arachidonic acid
in micelles by superoxide and hydrogen peroxide. J. Biol.
Chem. 256: 260-265 (1981). ' '

K.L. Fong, P.B. McCay, J.L. Poyer, H.P. Misra, and 8.8. Keele.
Evidence for superoxide-dependent reduction of Fe3 and its
role in enzyme-generated hydroxyl radical formation. Chem.
Biol. Inter. 15: 77-89 (1976). '

J.M.C. Gutteridge. Lipid peroxidation initiated by
superoxide-dependent hydroxyl radicals using complexed iron
and hydrogen peroxide. FEBS Lett. 172: 2u5-2u9 (198“).

W. Flitter, D.A. Rowley and B. Halliwell. Superoxide-
dependent formation of hydroxyl radicals in the presence of
iron salts. What is the physiological iron chelator? FEBS
Lett. 1§§:'31o-312 (1983).

J.M.C. Gutteridge. Ferrous ion-EDTA-stimulated phospholipid
peroxidation. A reaction changing from alkoxyl-radical to
hydroxyl-radical- dependent initiation. Biochem. J. 223:
697-701 (198“) ‘ '

M. Tien, L.A. Morehouse, J.R. Bucher and S.D. Aust. The
multiple effects of ethylenediaminetetracetate in several
model lipid peroxidation systems. Arch. Biochem. BiOphys.
218: H50-H58 (1983). '

J. M. McCord and E. D. Day, Jr. Superoxide-dependent production
of hydroxyl radical catalyzed by the iron-EDTA-complex. FEBS
Lett.86:139-1u2 (1978).

B. Halliwell. Superoxide-dependent formation of hydroxyl
radicals in the presence of iron chelates: Is it a mechanism

H9.

50.

51.

52.

53.

5“.

55.

56.

57.

58.

37

for -OH radical in biological systems? FEBS Lett. 92: 321-326
(1978)

R.A. Floyd. Direct demonstration that ferrous ion complexes
of di- and triphosphate nucleotides catalyze hydroxyl free
radical formation from hydrogen peroxide. Arch. Biochem.
Biophys. 225: 263-270 (1983). ' '

R. A. Floyd and C. A. Lewis. Hydroxyl free radical formation
from hydrogen peroxide by ferrous iron-nucleotide complexes.
Biochem. 22: 2645-26u9 (1983).

A.C.M. Filho, M.E. Hoffman and R. Meneghini. Cell killing and
DNA damage by hydrogen peroxide are mediated by intracellular
iron. Biochem. J. 218: 273-275 (198A).

T.C. Pederson and S.D. Aust. The mechanism of liver
microsomal lipid peroxidation. Biochim. Biophys. Acta 385:
232-2u1 (1975). ' '

L.A. Morehouse, M. Tien, J.R. Bucher and S.D. Aust. Effect of
hydrogen peroxide on the initiation of microsomal lipid
peroxidation. Biochem. Pharmacol. 32: 123-127 (1983).

L.A. Morehouse, C.E. Thomas and S.D. Aust. Superoxide
generation by NADPH-cytochrome PHSO reductase: The effect of
iron chelators and the role of superoxide in microsomal lipid
peroxidation. Arch. Biochem. Biophys. 232: 366-377 (198”)

J.R. Bucher, M. Tien, L.A. Morehouse and S.D. Aust. Three
mechanisms for the formation of an initiator of lipid
peroxidation by xanthine oxidase. In: Oxy Radicals and Their

 

Scavenger Systems, Volume I: Molecular Aspects (G. Cohen and

 

R.A. Greenwald, eds.). Pp. 296-299. Elsevier Science
Publishing Co., New'York (1983).

T.C. Pederson, J.A. Buege and S.D. Aust. Microsomal electron
transport. The role of reduced'nicotinamide adenine
dinucleotide phosphate-cytochrome c reductase in liver
microsomal lipid peroxidation. J. Biol. Chem. 238: 7131-7111
(1973). - ‘ - - - -

J.R. Bucher, M. Tien, L.A. Morehouse and S.D. Aust. Redox
cycling and lipid peroxidation: The central role of iron
chelates. Fund. Appl. Toxicol. 3: 222-226 (1983).

J.R. Bucher, M. Tien, L.A. Morehouse and S.D. Aust. Influence
of superoxide dismutase and catalase on strong oxidant
formation drug autoxidation of ferrous chelates. In: Ox
Radicals and their Scavenger Systems. Volume 1: Molecular
Aspects (G. Cohen and R.A. Greenwald. eds.). DD. 292-295.
Elsevier Science Publishing Co., New York‘(1983).

59.

6o.

61.

62.

63.

6H.

65.

66.

67.

68.

69.

70.

71.

38

J.R. Bucher, M. Tien and S.D. Aust. The requirement for
ferric in the initiation of lipid peroxidation by chelated
ferrous iron. Biochem. Biophys. Res. Comm. 111: 777-784
(1983). ' ' ' "'

G. Minotti and S.D. Aust. The requirement for ferric in the
initiation of lipid peroxidation by ferrous and hydrogen
peroxide (Submitted).

R. J. Youngman. Oxygen activation: is the hydroxyl radical
always biologically relevant? Trends in Biochem. Sci. 9:
280-283 (1989)

R.J. Youngman and E.F. Elstner. Oxygen species in paraquat
toxicity: The crypto-OH radical. FEBS Lett. 129: 265-268
(1981). ' ' °

A. Samuni, M. Chevion and G. Czapski. Unusual copper-induced
sensitization of the biological damages due to superoxide
radicals. J. Biol. Chem. 256: 12632-12635 (1981).

E.K. Hodgson and I. Fridovich. The interaction of bovine
erythrocyte superoxide dismutase with hydrogen peroxide:
Inactivation of the enzyme. Biochem. 12‘ 5299-5299 (1975).

G. E. Shull and E. C. Theil. Regulation of ferritin mRNA: A
possible gene-sparing phenomenon. J. Biol. Chem. 258:
7921-7923 (1983) ' '

J.B. Nielands. Microbial iron compounds. In: Annual Review
of Biochemistry, Volume 50 (E. E. Snell, P.D. Boyer, A. Meister
and C. C. Richardson, eds. ), pp. 715-731, Annual Reviews Inc.,
Palo Alto, CA (1981). '

 

R. G. Sheehan and E. P. Frenkel. The control of iron absorption
by the gastrointestinal mucosal cell. J. Clin. Invest. 51:
221-231 (1972). - - - -

R.R. Crosby. The control of iron balance by the intestinal
mucosa. Blood 22: uu1-uu6 (1963).

R.A. Muir, U. Hopfer and M. King. Iron transport across
brush-border membranes from normal and iron-deficient mouse
upper small intestine. J. Biol. Chem. 259: 8896-9903 (198”).

R.W. Topham, M.C. Walker and M.P. Calisch. Liver xanthine
dehydrogenase and iron mobilization. Biochem. Biophys. Res.
Comm. 109: 12HO-12fl6 (1982). ‘ ' ' ‘

M. N. Pearson and M. B Reich. Studies of ferritin and a new
iron-binding protein found in the intestinal mucosa of the rat.
J. Nutr. 99: 137- -1uo (1969).

72.

73.

79.

75.

76.

77.

78.

79.

80.

81.

82.

83.

8H.

39

M. Boulard, M. Delin and T.C. NaJean. Identification and
purification of a new non-heme, non-ferritin iron protein.
Proc. Soc. EXptl. Biol. Med. 139: 1379-138“ (1972).

M.C. Linder and H.N. Munro. The mechanism of iron absorption
and its regulation.‘ Fed. Proc. 36: 2017-2023 (1977).

J.W. Halliday, L. W. Powell and V. Mack. Iron absorption in
the‘rat: The search for possible intestinal mucosol carriers.
Br. J. Haematol. 3M: 237-250 (1976).

P. Aisen. Chemistry and physiology of the transferrins. In:
The Biological Chemistry of Iron (H.B. Dunford, D. Dolphin,
K.N. Raymond and L. Sieter, eds.). pp. 63-83, D. Reidel
Publishing Co., Dordrecht, The Netherlands (1982).

G. W. Bates and M. R. Schlabach. The nonspecific binding of
F93+ to transferrin in the absence of synergistic anions. J.
Biol. Chem. 250: 2177-2181 (1975L

M. R. Schlabach and G. W. Bates. The synergistic binding of
anions and Fe3+ by transferrin. J. Biol. Chem. 250: 2182-2188
(1975L ' ‘ " ‘ ‘

N. KoJima and G.W. Bates. The formation of Fe3*-transferrin-
0023 via the binding and oxidation of Fe2*. J. Biol. Chem.

ggg: 1203u-12039 (1981).

S. Osaki, D.A. Johnson and E. Frieden. The possible
significance of the ferrous oxidase activity of ceru10plasmin
in normal human serum. J. Biol. Chem. 281: 27M6-2751 (1966).

D.A. Baldwin, E.R. Jenny and P. Aisen. The effect of human
serum transferrin'and milk lactoferrin on hydroxyl radical
formation from superoxide and hydrogen peroxide. J. Biol.
Chem. 259:13391-1339“ (198A). ‘ '

J.M.C. Gutteridge, S.K. Paterson, A.N. Segal and B. Halliwell.
Inhibition of lipid peroxidation by the iron-binding protein
lactoferrin. Biochem. J. 199: 259-261 (1981).

J.V. Bannister, P. Bellavite, A. Davioli, P.J. Thornally and
F. Rossi. The generation of hydroxyl radicals following
superoxide production by neutrOphil NADPH oxidase. FEBS Lett.
150: 300- 302 (1982). '

N. Motohashi and J. Mori. Superoxide-dependent formation of
hydroxyl radical catalyzed by transferrin. FEBS Lett. 157:
197-199 (1983L “

J.M.C. Gutteridge. Antioxidant properties of the proteins
caeruloplasmin, albumin and transferrin. A study of their
activity in serum and synovial fluid from patients with

85.

86.

87.

88.

89.

90.

91.

92.
93.
9“.
95.

96.

97.

1

40 \

rheumatoid arthritis. Biochim. BiOphys. Acta. 869: 119-127 \\-

J.M.C. Gutteridge, S.K. Paterson, A.N. Segal and B. Halliwell.
Inhibition of lipid peroxidation by the iron-binding protein
lactoferrin. Biochem. J. 199: 259-261 (1981).

S. Young and A. Bomford. Transferrin and cellular iron
exchange. Clin. Sci. 61: 273-278 (198“).

R.D. Klausner, J.V. Renswonde, G. Ashwell, C. Kempf, A.N.
Schechter, A. Dean and R. R. Bridges. Receptor-mediated
endocytosis of transferrin in K562 cells. J. Biol. Chem. 258:
1715-172u (1983). ' ' '

N. Kojima and G.R. Bates. The reduction and release of iron
from Fe3*-transferrin-C023. J. Biol. Chem. 25”: 88h7-885u
(1979). ' ' ' '

E.R. Morgan. Studies on the mechanism of iron release from
transferrin. Biochim. Biophys. Acta 580: 312-326 (1979).

8.6. Kailis and E.R. Morgan. Iron uptake by immature
erythroid cells. Mechanism of dependence on metabolic energy.
Biochim. Biophys. Acta uou. 389-398 (1977).

A. Jacobs. Low molecular weight intracellular iron transport
compounds. Blood 50: 133-u39 (1977).

G.R. Bartlett. Phosphate compounds in rat erythrocytes and
reticulocytes. Biochem. Biophys. Res. Comm. 19: 1055-1062
(1976). - - - - ' - -

S. Pollack and T. Campana. Low molecular weight non-heme iron
and a highly labeled heme pool in the reticulocyte. Blood 59:
564-566 (1980).

C. G. D. Morley and A. Bezkorovainy. Identification of the iron
chelate in hepatocyte cytosol. IRCS Med. Sci. 11: 1106-1107
(1983). . . . .. .. ..

P.M. Harrison. Ferritin: An iron-storage molecule.' Sem.
Hematol. 13: 55-70 (1977).

P. Aisen and I. Listowsky. Iron transport and storage
proteins. In:' Annual Review of Biochemistry, Volume R9 (E.E.
Snell, P.D. Boyer, A. Meister and C.C. Richardson, eds.). PP.’
357-393, Annual Reviews, Inc., Palo Alto, CA (1980). ‘

 

D.R. Rice, C.C. Ford, J.L. White, J.M.A. Smith and P.M.
Harrison. The spatial structure of horse spleen apoferritin.
In: Iron Binding Proteins Without Cofactors or Sulfur

98.

99.

100.

101.

102.

103.

10".

105.

106.

107.

108.

109.

110.

41

Clusters (E. C. Theil. G. L. Eichhorn and L. G. Marzilli, eds. ),
pp. 39 -50, Elsevier, New York (1983).

E.C. Theil. Ferritin: Structure, function, and regulation.
In: Iron Binding Proteins Without Cofactors or Sulfur
Clusters (E.C. Theil. G.L. Eichhorn and L.G. Marzilli, eds.).

pp. 1-38, Elsevier, New York (1983).

M. Wauters, M. Michelson and R. R. Crichton. Studies on the
mechanism of ferritin formation.' FEBS Lett. 21: 276-280
(1978). ’ ' '

G. Melino, S. Stefanini, E. Chiancone and E. Antonini.
Stoichiometry of iron oxidation by apoferritin. FEBS Lett.
86: 136-138 (1978). '

D. E. Mayer, J. S. Rohrer, D. A. Schoeller and D. C. Harris. Fate
of oxygen during ferritin iron incorporation. Biochemistry
22: 876-880 (1983)

J.G. Wardeska, B. Viglione and N.D. Chasteen. Metal ion
complexes of apoferritin. Evidence for the initial binding in
the hydrophilic channels. J. Biol. Chem. 261: 6677-6683
(1986). ' ' '

P.M. Harrison, T.C. Hoy, 1.0. Macara and R. J. Hoare. Ferritin
iron uptake and release. Structure-function relationships.
Biochem. J. 1N3: HHS-#51 (197k).

C.F.A. Bryce and R.R. Crichton. The catalytic activity of
horse spleen apoferritin. Preliminary kinetic studies and the
effect of chemical modification. Biochem. J. 133: 301-309.

T. Jones, R. Spencer and C. Walsh. Mechanism and kinetics of
iron release from ferritin by dihydroflavins and dihydroflavin
analogues. Biochemistry 11: “011-fl017 (1978).

A. Mazur, S. Baez and E. Shorr. The mechanism of iron release
from ferritin as related to its biological properties. J.
Biol. Chem. 213. 1u7- 160 (1955) ‘

S. Granick and L. Michaelis. Ferritin II. Apoferritin of
horse spleen. J. Biol. Chem. 1H7: 91-97 (1993).

R.R. Crichton. Ferritin. Struct. Bond 11: 67-13" (1973).

R. R. Crichton, F. Roman and F. Roland. Iron mobilization from
ferritin by chelating agents.‘ J. Inorg. Biochem. 13: 305-316
(1980). - - -- - -

S. Sirivech, J. Driskell and B. Frieden. NADH-FMN
oxidoreductase activity and iron content of organs from
riboflavin- and iron-deficient rats. J. Nutr. 107: 739-7fl5
(1977) '

111.

112.

113.

11".

115.

116.

117.

118.

119.
120.
121.

122.

123.

42

Z. Zaman and R.L. Verwilghen. Effect of riboflavin deficiency
on activity of NADH-FMN oxidoreductase (ferriductase) and iron
content of rat liver. Biochem. Soc. Trans. 2: 306-308 (1977).

A. Mazur, S. Green, A. Saha and A. Carleton. Mechanism of
release of ferritin iron in vivo by xanthine oxidase. J.
Clin. Invest. 37: 1809- 1817 (1958). '

A. Mazur and A. Carleton. Hepatic xanthine oxidase and
ferritin iron in the developing rat. Blood 26: 317-322
(1965). ‘ '

G. D. Watt, R. B. Frankel and C.C. Papaefthymiou. Reduction of
mammalian ferritin.‘ Proceed. Natl. Acad. Sci.'82: 36u0-36u3
(1985). ‘ ‘ ‘ '

R. J. Ulvik. Relevance of ferritin-binding sites on isolated
mitochondria to the mobilization of iron from ferritin.
Biochim. Biophys. Acta 715: M2-51 (1982).

R.J. Ulvik and I. Romslo. Reduction of exogenous FMN by
isolated rat liver mitochondria. Significance to the
mobilization of iron from ferritin. Biochim. Biophys. Acta

332. 1157-1169 (1981).

J.M.C. Gutteridge, B. Halliwell. A. Treffry, P.M. Harrison and
D. Blake. Effect of ferritin-containing fractions with
different iron loading on lipid peroxidation. Biochem. J.
209: 557-560 (1983). ~ '

M. O'Connell, B. Halliwell, C.P. Moorhouse, 0.1. Aruoma, H.
Baum and T.J. Peters. Formation of hydroxyl radicals in the
presence of ferritin and haemosiderin. Biochem. J. 235:
727-731 (1986). ' '

E.D. Wills. Mechanisms of lipid peroxide formation in animal
tissues. Biochem. J. 29: 667-676 (1966).

S.M.H. Sadrzadeh, E. Graf, S. S. Panter, P. E. Hallaway and J.W.
Eaton. Hemoglobin. A biologic Fenton reagent. J. Biol.
Chem. 259: 1H358-19356 (198”). '

J. Kanner and S. Harel. Initiation of membranal lipid
peroxidation by activated metmyoglobin and methemoglobin.
Arch. Biochem. Biophys. 237: 311-321 (1985).

C. C. Hinterbourn. Hydroxyl radical production in body fluids.
Roles of metal ions, ascorbate, and superoxide. Biochem. J.
198: 125-131 (1981). " ‘

J.M.C. Gutteridge, D.A. Rowley and B. Halliwell.
Superoxide-dependent'formation of hydroxyl radicals and lipid

12“.

125.

126.

127.

128.

129.

130.

131.

132.

133.

139.

43

peroxidation in the presence of iron salts. Biochem. J. 206:
605-609 (1982). ' '

J. M. McCord. Free radicals and inflammation: Protection of
synovial fluid by superoxide dismutase. Science 185: 529-531
(197”) ' ’

B. Halliwell. Superoxide-dependent formation of hydroxyl
radicals in the presence of iron salts. Its role in
degradation of hyaluronic acid by a superoxide-generating
system. FEBS Lett. 26: 238-2N2 (1978).

D. Rowley, J.M.C. Gutteridge, D. Blake, M. Farr and B.
Halliwell. Lipid peroxidation in rheumatoid arthritis:
thiobarbituric acid-reactive material and catalytic iron salts
in synovial fluid from rheumatoid patients. Clin. Sci. 66:
691-695 (198“). ' '

J.S. Bus, S.D. Aust and J.F. Gibson. Superoxide- and singlet
oxygen-catalyzed lipid peroxidation as a possible mechanism
for paraquat (methyl viologen) toxicity. Biochem. BiOphys.
Res. Comm. 58: 799-755 (197“). ' '

M.A. Trush, E.G. Mimnaugh, E. Ginsburg and T.E. Gram. In
vitro stimulation by paraquat of reactive oxygen-mediated
lipid peroxidation in rat lung microsomes. Toxicol Appl.
Pharmacol. 60: 279-286 (1981). ‘

D. J. Kornbrust and R. D. Mavis. The effect of paraquat on
microsomal lipid peroxidation in vitro and in vivo. Toxicol.
Appl. Pharmacol. 53: 323- 332 (1980). '

H. P. Misra and L. D. Gorsky. Paraquat and NADPH-dependent
lipid peroxidation in lung microsomes. J. Biol. Chem. 256:
999h-9998 (1981). ' ' '

L.L. Smith, M.S. Rose and I. Hyatt. The pathology and
biochemistry of paraquat. In: Oxygen Free Radicals in Tissue
Damage (Ciba Foundation Symposuim 65) pp. 321-3fl1, Excerpta
Medica, New York (1979).

R. Kohen and M. Chevion. Paraquat toxicity is enhanced by
iron and reduced by desferrioxamine in laboratory mice.
Biochem. Pharmacol. 3H: 18M1-18u3 (1985).

R. KOhen and M. Chevion. Transition metals potentiate
paraquat toxicity. Free Rad. Res. Commun. 1: 79-88 (1985).

H.G. Sutton and C.C. Hinterbourn. Chelated iron-catalyzed jOH
fOrmation from paraquat radicals and R202: Mechanism 0f
formate oxidation. Arch. Biochem. Biophys. 235: 106-115
(198”). . . o o:

135.

136.

137.

138.

139.

1’40.

191.

1H2.

1N3.

1””.

It's.

196.

44

R.D. Fairshter, N.D. Vaziri, L.C. Dearden, K. Halott and H.
Caserio. Effect of dimethylthiourea on paraquat toxicity in
rats. Toxicol. Appl. Pharmacol. 13: 150-15” (198”).

A. C. Bagley, J. Krall and R. E. Lynch. Superoxide mediates the
toxicity of paraquat for Chinese hamster ovary cells. Proceed.
Natl. Acad. Sci. 83. 3189-3193 (1986).

A.P. Autor. Reduction of paraquat toxicity by superoxide
dismutase.’ Life Sci. 13: 1309-1319 (197").

C. E. Patterson and M. L. Rhodes. The effect of superoxide
dismutase on paraquat mortality in mice and rats. Toxicol.
Appl. Pharmacol. 62: 65-72 (1982). '

J. Goodman and P. Hochstein. Generation of free radicals and
lipid peroxidation by redox cycling of adriamycin and
daunomycin. Biochem. BiOphys. Res. Comm. 11: 797-803 (1977).

J.H. Doroshow, 0.1. Locker and C.E. Myers. Enzymatic defenses
of the mouse heart against reactive oxygen metabolites.l.
Alterations produced by doxorubicin. J. Clin. Invest.'§§:
128-135 (1980). ' '

E.G. Mimnaugh. T.E. Cram and M.A. Trush. Stimulation of mouse
heart and liver microsomal lipid peroxidation by anthracycline
anticancer drugs: Characterization and effects of reactive
oxygen scavengers. J. Pharmacol. Exper. Ther. 226: 806-816
(1983). ' ' ' ' ‘

L. Sterrenberg, R.H.M. Julicher, A. Bast and J. Noordhoek.
Adriamycin stimulates NADPH-dependent lipid peroxidation in
liver microsomes not only by enhancing the production of 02'
and H202, but also by potentiating the catalytic activity of
ferrous ions. Toxicol. Lett. 32: 153-159 (198“).

H. Muliwan, M.E. Scheulen and H. Kappus. Adriamycin
stimulates only the iron ion-induced, NADPH-dependent
microsomal alkane formation. Biochem. Pharmacol. 31:
31h7-3150 (1982). ' ' ‘

E.G. Mimnaugh, M.A. Trush and T.E. Gram. Enhancement of rat
heart microsomal lipid peroxidation following doxorubicin

treatment in vivo. Cancer Treat. Rep. 61: 731-733 (1983).

E.R. Herman, A.N. El-Hage, V.J. Ferrans and T. Hitiak.
Reduction by ICRF-187 of acute daunorubicin toxicity in Syrian
golden hamsters. Res. Comm. Chem. Pathol. Pharmacol. fig:
217-231 (1983).' ' ‘ ' '

K. Sugioka, H. Nakano. T. Noguchi, J. Tsuchiya and M. Nakano.
Decomposition of unsaturated phospholipid by iron-ADP-
adriamycin co-ordination complex. Biochem. BiOphys. Res.
Comm. 199: 1251-1258 (1981). ‘ ‘

197.

148.

199.

150.

151.

152.

153.

45

K. Sugioka and M. Nakano. Mechanism of phospholipid
peroxidation induced by ferric ion-ADP-adriamycin
co-ordination complex. Biochim. Biophys. Acta 713: 333-3fl3
(1982). ‘ ' ' ‘

K. Sugioka, H. Nakano, M. Nakano, S. Tero-Kubota and Y.
Ikegami. Generation of hydroxyl radicals during the enzymatic
reductions of the Fe31-ADP-phosphate-adriamycin and Fe3 -ADP-
EDTA systems. Biochim. Biophys. Acta.1§;: u11-uz1 (1983).

N. Nakano, K. Ogita, J.M.C. Gutteridge and M. Nakano.
Inhibition by the protein ceruloplasmin of lipid peroxidation
stimulated by an Fe3+-ADP-adriamycin complex. FEBS Lett. 166:
232-236 (198”). ' ’ '

D.A. Rowley and B. Halliwell. DNA damage by superoxide-
generating systems in relation to the mechanism of action of
the anti- tumor antibiotic adriamycin. Biochim. BiOphys. Acta
761: 86-93 (1983). ' '

H. Eliot, L. Gianni and C. Myers. Oxidative destruction of
DNA by the adriamycin-iron complex. Biochem. 23: 928-936
(198u). ‘

E. J. F. Demant. Transfer of ferritin-bound iron to adriamycin.
FEBS Lett. 176: 97— 100 (1981).

E. J. F. Demant and N. Norskov-Lavritsen. Binding of
transferrin-iron by adriamycin at acidic pH. FEBS Lett. 126:
321-32" (1986). ' '

CHAPTER I

FERRITIN AND SUPEROXIDE-DEPENDENT LIPID PEROXIDATION

46

ABSTRACT

Intracellular, low molecular weight iron complexes are proposed
to promote membrane lipid peroxidation. However, hepatic cytosol
prepared from iron-loaded rats contained little iron associated with
proteins of less than 10,000 daltons as Judged by gel filtration
chromatography and ultrafiltration studies. Nearly 70} of the total
cytosolic iron was found in ferritin, which promoted the peroxidation
of phospholipid liposomes when incubated with xanthine oxidase,
xanthine, and ADP. Activity was inhibited by SOD but markedly
stimulated by the addition of catalase. Xanthine oxidase-dependent
iron release from ferritin was also inhibited by SOD suggesting that
027 can mediate the reductive release of iron from ferritin. Catalase
had little effect on the rate of iron release from ferritin, thus H202
appears to inhibit lipid peroxidation by preventing the formation of
an initiating species rather than by inhibiting iron release from
ferritin. ESR spin trapping with DMPO was used to observe free
radical production in this system. Addition of ferritin to the
xanthine oxidase system resulted in loss of the 027 spin trap adduct
suggesting an interaction between 027 and ferritin. If DMPO was added
2 min after ferritin addition the resultant spectrum was that of a fOH
spin trap adduct which was abolished by the addition of catalase.
These data suggest that ferritin may function in Zilfl as a source of
iron for promotion of 027-dependent lipid peroxidation. Stimulation
of lipid peroxidation but inhibition of fOH formation by catalase
suggests that. in this system, initiation is not via an iron-catalyzed

~ Haber-Heiss reaction.

47

INTRODUCTION

Cellular damage resulting from ischemia (1), hyperoxia (2), redox
cycling (3). and other oxygen associated toxicities is often
attributed to enhanced production of 027 with biochemical alterations
reported to include peroxidation of membrane phospholipids (A) and DNA
degradation (5). However, in aqueous media 027 is relatively
unreactive towards most organic compounds (6) thus its prOposed
deleterious effects may be the result of its participation in
reactions leading to other more reactive species.

Superoxide-dependent formation of more reactive radicals such as fOH
requires the presence of transition metal ions such as copper or iron
(A). The most widely proposed mechanism is the iron-catalyzed

Haber-Weiss reaction (7):

 

2027 + 211* A, 02 + H202 (2)
Fe2+ + 11202— Fe3" + 0H“ + '01! (3)

Although 70H is highly reactive, 1t3.lfl.!£!2 formation is
contingent upon the availability of physiological iron. The ability
of both lactoferrin (8) and transferrin (9) to generate fOH in
the presence of a 027 generating system has been demonstrated.
Ninterbourn (10) and others (11) have reported the ability of iron in
extracellular fluids such as synovial and cerebrospinal fluid to
catalyze fOH formation and promote low rates of lipid peroxidation.

However, the ability of intracellular iron to participate in redox
48

49

reactions is relatively unknown although investigators have postulated
that cells may contain low molecular weight iron complexes capable of
promoting lipid peroxidation (u, 12). Nucleotide chelated iron
complexes have been isolated from erythrocytes (13, 1h) but definitive
identification of similar iron chelates in liver is lacking with
suggested chelators ranging from ADP (15) to citrate (16).

Although the existence of low molecular weight complexes remains
controversial, it is known that the maJority of intracellular iron is
stored within ferritin, a large multi-subunit protein found
predominately in the liver, spleen, and bone marrow (17). Ferric iron
(up to A,500 atoms/ferritin complex) is stored within the central core
of ferritin as ferric hydroxide complexed with phosphate (18). The
spherical ferritin complex contains a central core and six shallow
'pockets' through which iron is deposited or mobilized (19). While it
is accepted that mobilization of ferritin iron requires reduction to
the ferrous state (20), physiological reductant(s) remain unknown.
Release of iron from the core may necessitate passage of the reductant
through the narrow pockets, therefore potential biological reductants
may be limited by size constraints.

Reducing agents such as dithionite and thioglycolate readily
release ferritin iron in 11359 (21, 22) while reductants with
potential physiological significance (ascorbate, cysteine, and
GSH) do so much more slowly (23). Chelating agents (EDTA and
2,2'-dipyridyl) can facilitate iron release from ferritin but only in
the presence of a reductant (2A). The most effective mobilizers of
ferritin iron appear to be reduced flavins (25). Mazur (26,27)

originally proposed that xanthine oxidase, in its HAD-requiring

50

dehydrogenase form, may promote the release of iron from ferritin by
direct reduction of the ferric iron. More recent work by T0pham gt
21° (28) has also provided evidence that xanthine dehydrogenase may
participate in the mobilization of iron from ferritin.

The similarities in the conditions under which ferritin iron is
mobilized and lipid peroxidation occurs, i.e., the presence of
reducing equivalents and chelating agents, is striking. Nonetheless,
relatively few studies have investigated the ability of ferritin to
supply iron for the promotion of lipid peroxidation. Gutteridge (29)
reported that ferritin was able to facilitate in ZLEEQ
ascorbate-dependent lipid peroxidation. Similarily, Hills (30)
demonstrated that ferritin may promote non-enzymatic lipid
peroxidation, however, it was subsequently suggested that ferritin
iron was not responsible for the enhancement of lipid peroxidation
noted in microsomes isolated from iron-loaded rats (31).

Xanthine oxidase, which generates 027 and H202 during its
conversion of xanthine to urate (32), is often used for in litgg lipid
peroxidation studies as 027 readily reduces ferric iron. The small
size of 027, in conJunction with its ability to reduce chelated iron,
suggests it to be an excellent candidate for mobilization of iron from
ferritin. The existing controversy over the ability of xanthine
oxidase to directly reduce and release ferritin iron provides further
rationale for these studies.

In this paper, we report our findings which demonstrate that 027,
as generated by xanthine oxidase, reductively releases ferritin bound
iron. Once released this iron can promote the peroxidation of

phospholipid liposomes. Catalase markedly stimulates MDA formation in

51

this system, suggesting that initiation is not dependent upon H202.
These results were further supported by the use of ESR spin trapping
which demonstrated a negative correlation between fOH formation and

lipid peroxidation.

MATERIALS AND METHODS

Materials

Xanthine, cytochrome c (Type VI), 2-TBA, ADP, butylated
hydroxytoluene, mannitol, ”,7-diphenyl-1,10-phenanthroline, potassium
superoxide and crown ether were purchased from Sigma Chemical Company
(St. Louis, MO). Thioglycolate and 2,2'-dipyridyl were from Fisher
Scientific (Fairlawn, NJ). DMPO and dimethyl sulfoxide were obtained
from Aldrich Chemical Company (Milwaukee, WI) and H202 was a product
of Mallinckrodt Chemical (Paris, KY). Sephadex G-200, Sephadex G-25
and Sepharose 68 were from Pharmacia (Piscataway, NJ) while Ultragel
AcA nu was purchased from LKB (Bromma, Sweden). Imferon was a gift
from Merrell Dow Pharmaceuticals (Cincinnati, OH). DMPO was vacuum
distilled prior to use while all other chemicals were of analytical
grade or better and used without further purification. All buffers
and reagents were passed through Chelex 100 (Bio-Rad Laboratories)
ion-exchange resin to free them of contaminating transition metal
ions.
Enzymes

Bovine erythrocyte SOD (EC 1.15.1.1) and buttermilk xanthine
oxidase (EC 1.2.3.2) were obtained from Sigma Chemical Company.
Catalase (EC 1.11.1.6) was purchased from Millipore (Freehold, NJ).
Gel filtration chromatography on Sephadex G-25 was utilized to remove
the antioxidant thymol and ammonium sulfate from the commercial
preparations of catalase and xanthine oxidase, respectively. After
chromatography, xanthine oxidase activity was measured by aerobic
reduction of cytochrome c (33), with a unit of activity defined as 1

umol cytochrome c reduced/min/ml. Superoxide dismutase activity was

52

53

measured by the method of McCord and Fridovich (33) and catalase by
the procedure of Beers and Sizer (3H).
Fractionation of Rat Liver Cytosol

Male, Sprague—Dawley rats (275-300g) received 25 mg of Imferon
(iron-dextran) intraperitoneally 2“ h prior to sacrifice. Livers were
excised, perfused, and homogenized in an equal amount (w/v) of 50 mM
NaCl. The homogenate was centrifuged at 8,000 x g for 20 min and the
resulting supernatant was subsequently centrifuged at 105,000 x g for
90 min. The supernatant was carefully decanted and applied to an
Ultragel AcA HA gel filtration column (8 x 85 cm) and eluted with 50
mM NaCl. Following an initial elution of 1,500 ml, fractions (19 ml)
were collected and analyzed for total iron content. Five distinct
iron-containing peaks were revealed and the appropriate fractions
pooled for further analysis.

Rat hepatic cytosol, prepared from both control and iron loaded
rats as detailed above, was fractionated by ultrafiltration on an
Amicon ultrafiltration apparatus equipped with a PM 10 membrane. The
filtrate and the retained fractions were then analyzed for total iron
content.

Purification of Rat Liver Ferritin

Ferritin was purified from the livers of Imferon treated rats
essentially as described by Halliday (35). Male Sprague-Dawley rats
(250-300 3) received 12.5 mg of elemental iron as Imferon peritoneally
on alternate days for one week (50 mg total dose). Rats were
terminated by decapitation and livers were excised, perfused and
homogenized in two volumes of distilled water. The homogenate was

quickly heated to 70°C with continuous stirring for 5 minutes,

5A

and centrifuged at 1,

ed to N°C,

ediately cool
protein.

imm
tate denatured

to precipi
nium sulfate and allowed to

aturated ammo
,500 x g

ated by 50$ 3
gation at 1

after centrifu
0.02 M phosph
zed for 2“ ho

tin precipit

ferri
ate buffer,

stir overnight.
then resuspe
urs

0 minutes was
de) and dialy

for 3
pH 7.“, conta
t A liters of

odium chlori
After

fer.

ining 0.1 M 3
change of buf

for 30 minutes.

30,000 x g

agains
g for 2

dialysis the s
n centrifuged

tant supernatant was the

The resul
se 68 (2.5 x
e ferritin pe id volume and

n spectrophot
The combi

on Sepharo
d near the vo

eadily visibl
oled based 0
iron analysis.

ak was elute

The r
s were then

fractions po
ned fraction

PM 30 membrane.

0-200 column (1.
then poole
acrylami de gel .

Prior to use,

9011
entration of

o a final conc
n was loosely

ferritin t
hour. This was done to ensure that no iro
the protein. Subsequent chromatography on Sephadex 0-25 before use
d the EDTA and ferritin.

vely separate

effecti

Preparation of Microsomal Lipid and Liposomes
ley rats (250-275 g) were obtained from Charles

ch liver microsomes were isolated as per

55

isolated microsomes by the method of Folch gt 3;. (37). All solvents
utilized were purged with argon and all steps performed at H°C to
minimize autoxidation of unsaturated lipids. Phospholipid liposomes
were subsequently prepared by indirect, anaerobic sonication (38) and
lipid phosphate assays performed according to Bartlett (39).

Lipid Peroxidation Assays

 

Xanthine oxidase dependent peroxidation of liposomes was
performed by incubating liposomes (1 umol lipid phosphate/ml) with
0.33 mM xanthine and xanthine oxidase and rat liver ferritin as
specified in figure legends. Reaction mixtures were constituted in
50 mM NaCl, pH 7.0, and incubated at 37°C in a Dubnoff metabolic
shaker bath under an air atmosphere. Although unbuffered, incubations
remained at pH 7.0 throughout the course of the experiments.
Peroxidation was monitored by taking aliquots from the incubations at
specified times and measuring MDA formation using the TBA test (10).
Butylated hydroxytoluene (0.03 volumes in 21 ethenol) was added to the
thiobarbituric acid reagent to prevent further peroxidation of lipid
during the assay procedure.

Assays for Total Iron and Ferritin Iron Release

 

Total iron was determined essentially as described by Brumby and
Massey (A1). Aliquots of sample, or citrate-iron (8.5:0.5 mM) for a
' standard curve, were brought to 0.15 ml with water. To each sample
was added 0.05 ml of thioglycolic acid (10% v/v) and 0.2 ml of glacial
acetic acid followed by vigorous mixing. After allowing 30 minutes
for protein denaturation, 0.32 ml of water, 0.28 ml of saturated
sodium acetate, and 1 ml of 2.0 mM “,7-diphenyl-1,10-phenanthroline in

isoamyl alcohol were added. The mixture was agitated, centrifuged in

56

an IEC table tap centrifuge and the absorbance of the upper pink layer
read at 535 nm. Care was taken to extract endogenous iron from the
saturated sodium acetate with the phenanthroline/isoamyl alcohol
mixture prior to use.

Measurement of iron release from ferritin was performed according
to Mazur 32 al. (27) with modification, relying upon absorbance of the
ferrous-dipyridyl chromOphore. Incubations in a final volume of 1 ml
contained 0.025 units of xanthine oxidase, 0.33 mM xanthine, 500 uM
ferritin iron, and 5.12 mM 2,2'-dipyridyl in 50 mM NaCl, pH 7.0. In
the experiments involving potassium superoxide, xanthine oxidase and
xanthine were replaced with 2 mM potassium superoxide prepared in
crown ether essentially as described by Ruddock 33 al. (AZ). Catalase
and SOD were included as indicated in figure legends. Reactions were
continuously monitored at 520 nm in a Cary 219 dual beam recording
spectrophotometer. The amount of iron released from ferritin was
determined from a standard curve using ferric chloride reduced with
0.1 ml of 101 thioglycolate.

ESR Spin Trapping

 

Detection of radicals produced by xanthine oxidase in the
presence or absence of ferritin was accomplished by ESR spin trapping
experiments using 60 mM DMPO as outlined in the figure legends.
Incubations were constituted and directly transferred to the cuvette
of a Varian Century-112 EPR spectrometer. Spectrometer settings were:
3329.” G magnetic field, 15 mw microwave power, 9.h232 GHz, 1000 KHz
modulation frequency, 0.63 modulation amplitude, 2.0 second time

constant and 8 minute scan time.

RESULTS

Fractionation Of Rat Liver Cytosol

The data in Table 1 demonstrate that 98.8% of the total cystolic
iron was recovered in association with proteins in the range of
molecular weight corresponding to 500-100 kdaltons. No iron was
detected in the region expected to yield a low molecular weight form
of iron (< 5,000 daltons). Similarly, ultrafiltration of rat liver
cytosol frOm both control and iron-loaded rats confirmed that less
than 1} of the total iron in the homogenates was recovered in the
filtrate (< 10,000 daltons) and 6A1 in the retained fraction
(> 10,000 daltons) (Table 2). The remainder of the homogenate iron
was recovered in the pellets obtained from the centrifugation at

8,000 x g and 105,000 x g.

Table 2. Total Iron Analysis of Rat Liver Homogenate and
Ultrafiltrated Cytosol

 

nmol Fe3+lml (Range)

 

Fraction Control Iron loaded
Homogenate - 6u0-880 1600-1920
Retained (> 10 kdaltons H10-560 ‘8“0-870
Filtrate (< 10 kdaltons) 8-8 5-6

 

Livers from control and iron-loaded rats (12.5 mg Imferon, 12 hr.
prior to sacrifice) were homogenized in 50 mM NaCl and cytosol
prepared as indicated in Materials and Methods. The cytosol (10 ml)
was ultrafiltrated using a PM 10 membrane to near dryness and the
volume adjusted to 10 ml with Chelex-treated 50 mM NaCl. Aliquots of
the homogenate, the filtrate, and the retained fractions were then
analyzed for total iron content as described in Materials and Methods.
Data given are for two animals in each group.

57

58

.ncouamux cowboom no omens unmaoz Luasooaoa m wcqgo>oo

mcofiuomsu o>Hu coca uoHooq can con" Hmuo» Lou oouhamcm one nouooaaoo one: acoauomnm
.0.» ma .Homz,:a cm so“: ago mm x mv cssaoo a: <04 Homaguaa am so umnamgmoumaogno
mm: .moocuoz cam namfinoumz ca omnfinommu mm oopmaonn .Aaa mmpv Honouho Lo>ua umm

 

0.00 mmw mamuoa

 

 

0pm 0 0

0pm 0

0.00 em

0V00 =0P 000.000
ooaaanm oaaemm.amuou no a Hax+mom Hoe: owcmm I:

 

Homoumo Lo>aq
umm no anqumOpmeoLno Bosh nocamuno ncofiuomnm omaoom no mamaamc< :oLH Hmuop .p magma

59

The fraction containing nearly 70% of the cytosolic iron eluted
in the ”00'500 kdalton range. It was subsequently determined by SDS
gel electrophoresis and Ouchterlony double diffusion analysis that
this iron was associated with ferritin (results not shown).

Xanthine Oxidase-Dependent Lipid Peroxidation

As shown in Figure 1, MDA formation was observed in a lipid
peroxidation system employing liposomes, purified rat liver ferritin,
and xanthine oxidase. Increasing the amount of xanthine oxidase
beyond 0.005 U/ml to generate greater rates of 027 production had an
inhibitory effect on MDA formation. The addition of catalase (10 and
25 U/ml) markedly increased the rate of lipid peroxidation observed at
all xanthine oxidase concentrations used (Figure 1). The maximum rate
of MDA formation occurred at greater xanthine oxidase activity as the
concentration of catalase was increased with the maxima occurring at
0.015 and 0.02 U/ml of xanthine oxidase with 10 and 25 U/ml of
catalase, respectively. Similarly, overall rates of lipid
peroxidation increased with the shift to higher activities of the two
enzymes. In the absence of xanthine oxidase or ferritin rates were 22
and 301 of the complete system in the absence of catalase. The
inclusion of catalase had no effect on these rates (data not shown).

When xanthine oxidase concentration was held constant at 0.005
U/ml, where the rate of MDA formation was maximal in the absence of
catalase, increasing concentrations of catalase up to 100 U/ml
resulted in increased rates of MDA formation as shown in Figure 2.

To determine whether enhancement of lipid peroxidation by

catalase was related to the use of ferritin as the source of iron, the

6O

.ucmucoo <0: Lou omzmmmm one: mogsuxwa mcofiuommn

on» sock muozcfiam cam ommofixo meanucmx Mo coauaoum on» hp oopmfiuacq mm: :oHumuonamm

.o. 0 ma .Homz za om cw ommudxo magnucmx mo mcoapmpucmocoo mcahgm> cam Aasxa oopv

000H0000 .Aza mm. 00 mcfinscmx .Aza F0 000 .A .mmm a: 00m0 00000000 00>00 000 .Aaa\000000000
oHaHH Hos: Fv mmaomoaaa oHaHHonqmona cocamucoo Aoasao> Hanan as my mogsuxfia cofiuomom
.nuaowaosamonm no coaumuaxonom ocoocoaoosommoaxo moanucmx ocm cauqnnom .F ogsmam

61

 

no.0

:E\3 wm<o_xo mz._I._.Z<x
3.0 8.0 No.0 5.0

 

.

 

2.5: mm + «[0

_E\_.. 0.

1.!

 

 

.+.o.|o.

3223 o: I

n p n -

 

 

o
0'

O
0!
'Iw/Lcfiw/vow Iowu

Omd

.ucmncoo «a:

non omzmmmm ego: monsuxns conuommn on» Bonn muo=UHHm ocm ommonxo cannucmx no scannuom on»
no uoumnpncn mm: conumUonnmm .o.» ma .Homz zs om en ommamumo no mconumnpceocoo

manngm> 000 Aaa\= m00.00 mmmenxo mcnn0cmx .Aze mm.00 mcnnscmx .Azs _0 000 .A+mma

:3 oowv enunnnon Lo>na umn .Aaa\mumcamona Unana Hoe: nv mosomoana onanaonamona nmcnmucoo
Amasao> Hmcnn as my monspxna acnnomom .mOHQHHosamonm no conumonxonmm unoocoaoaimmmcnxo
mansucmx new anunnnmm co m:0numnucoocoo mamamumo mcnhnm> no uoonnm one .m onsmnn

63

2.53 mmdqflrdo

on. 00.

On

 

 

 

 

 

00.0

9.0

9.0

 

 

Im/U!W/VGW IOWU

64

effect of catalase was assessed when iron was supplied directly as
ADP-Fe3*. As shown in Figure 3 catalase (100 U/ml) was found to
stimulate the rate of MDA formation approximately two-fold at a
xanthine oxidase concentration of 0.01 U/ml.
Effect of SOD and Mannitol

The stimulation of MDA formation by catalase suggested that
initiation of lipid peroxidation in this system is inhibited by H202.
To further characterize lipid peroxidation in this system, the effect
of SOD and mannitol was also investigated. As can be seen in Table 3
SOD significantly inhibited MDA formation both in the presence and
absence of catalase while mannitol stimulated activity slightly. The
omission of xanthine oxidase or ferritin resulted in rates of lipid
peroxidation being 23 and 18% of control in the absence of catalase,
with the addition of catalase having no effect on these incubations.

Table 3. The Effect of SOD and Mannitol on Xanthine Oxidase-
Dependent Peroxidation of Phospholipids

 

nmol MDA/min/ml

 

- catalase + catalase
no additions 0.038 0.H80
+ SOD 0.007 0.015
+ mannitol 0.0h4 0.510
- xanthine oxidase 0.009 -"'-

- ferritin 0.007 """

 

Reaction mixtures (5 ml final volume) contained phospholipid liposomes
(1 nmol lipid phosphate/ml), rat liver ferritin (200 uM Fe3*), ADP (1
mM), xanthine (0.33 mM) and xanthine oxidase (0.0025 U/ml) in 50 mM ‘
NaCl, pH 7.0. Where indicated, reaction mixtures contained SOD (100
U/ml), mannitol (10 mM) or catalase (200 U/ml). Peroxidation was'
initiated by the addition of xanthine oxidase and aliquots from the
reaction mixtures were assayed for MDA content.

65

.ucmncoo 4oz Lon commmmm one:

monsuxna conuommn on» sonn mposcnam ocm ammonxo mannpcmx no :onpnoum on» no omnmnuncn mm:
conumunxoama .0.» m0 .0002 zs.0m an Aaa\=.00.0.000H0000 000 AHa\0 n0.00 mmmenxo mensucmx
.Aza mm.00 mcnsscmx .Aon :0 0n 000 m .m ._ 00 Fumv +m00100< .Aaa\000000000 enand H000 _0
moaomquH Unanaosqmona oocnmucoo Aossao> Hanan as my monsuxns connomom .mUnanaonqmocm no
cenmeonnmm acoucmamotmmmcnxo mannucmx ocm +mmn1mo< co mmmamnmo no aownnm one .m onsmnn

66

 

m

3233-

3223...

:2 3 Tnmm unadu—
N. m

u

m.
A

 

 

<1: N.
o o
lw/UW/VGW low”.

‘9
0

co
0

67

Release of Iron From Ferritin

Inhibition of lipid peroxidation by SOD in this system suggests
that 027 may function in the reductive release of iron from ferritin.
To test this hypothesis, iron release from ferritin was directly
measured with 2,2'-dipyridyl which readily complexes with ferrous
iron, forming a chromophore absorbing maximally at 520 nm.

As shown in Figure A, iron release from ferritin was strictly
dependent upon xanthine oxidase. Greater rates of iron release 6
occurred when the amount of ferritin added to the system was increased
(results not shown). Importantly, SOD completely inhibited the
release of iron from ferritin. To further confirm that 027 can
reductively release ferritin iron, potassium superoxide in crown ether
was found to also result in the release of iron from ferritin (Figure
5). Again, addition of SOD resulted in a decreased rate of iron
release.

Inclusion of catalase in either the xanthine oxidase or potassium
superoxide system resulted in an apparent stimulation of iron release.
The results in Figure h show that increasing concentrations of
catalase cause an apparent increase in the rate of iron release with
1000 U/ml of catalase apparently effectively scavenging all 8202
produced by xanthine oxidase (0.025 U/ml) or 027 dismutation as 2000
U/ml of catalase gave virtually identical rates.

ESR Spin Trapping

 

ESR spin trapping was utilized to observe radical production in
this system. This technique has been shown to be effective for
trapping of both 027 and fOH using DMPO as the radical trap (33). In

the absence of ferritin, xanthine oxidase generated 02? which reacted

.Amncsoao wennnm>v oooamuoo mafia o .A: copy now mafia—u .oc0nnnuuo o: o .oooonxo ocnnucox
manna 4 A .s: omm um naososcnucoo uonouncos vco oomonxo ocnnncmx no conuncom on» no
ooumnpncH ono: ozonpomom .vouoOnocn om ooosaocn ono: mom can ommaouoo .0.» ma .Hooz :8
om an A: m~0.00 ommunxo ocnspcmx 000 Axe mm.00 oensscmx .Aza mn.mv anennn0n01.m.m .A+m0n
z: oomv caunnnon no>na non nocnmpcoo Aossao> Hogan Ha Pv monsuxns conuooom .cnnnnnon
sonn ooooaom conH ucoucoqoouomoUon ouncucmx :o oooamumo cam now no uoonnm one .= onsmnn

69

 

EEV MSE.

 

 

 

 

 

m. o. m
# GXI.“ 111.-41:1! \lMuwl‘
Dom +0.1
3283 I
. zoom
noon \
3000. \\
=Goom .

lw/ pesoaplaj lowu

 

 

7O

.0.e m0

oomv anunnnon no>nH non oonnonnoo Aossao> Honnn as «v monsuxns :Onuomom
oomoaom nonH unoonoaoaooonxonoasm ssnommnon co ommaouou new now no noonnm one

.0000H00H one: A: 00.0 000 000 A: 000_0 000H0000 coumonecn mnonx
.Homz :0 0m an Axe NV wenxonoasm asnmnmpoa 000..Aza mn.mv anunnn0n01.~.m .n+m0n :0

.cnunnnon sonn
.m onsmnn

71

oow+

flab—Oncol

3228+

2:5 m2;
0.

N

 

 

 

Iw/pasoapl a :1 |OllJU

72

with DMPO yielding the DMPO-OOH adduct signal (hfs constants AN - 13.1

G, A H - 11.0 G and A H - 1.3 0) (Figure 6). Addition of ADP to

13 'r
the incubation had no effect on the intensity of the signal while
catalase appeared to prevent the decay of the signal adduct.

Formation of the DMPO-OOH adduct was dependent upon enzymatic activity
as no signal was observed in the absence of xanthine.

When ferritin was present in this system no DMPO~00H signal was
observed, suggesting that 027 was interacting with the ferritin
molecule (Figure 7A). When DMPO was added two minutes after xanthine
oxidase additions, DMPO-OH (hfs constants AN - AH - 1u.8 0) but no
DMPO-00R signal adduct was observed (Figure 7B). The inclusion of ADP
had little effect on the DMPO-OH signal intensity. Catalase (200
U/ml) prevented DMPO-OH adduct formation in this system, indicating

that fOH formation required H202.

73

Figure 6. ESR Spin Trapping of Superoxide Generated by Xanthine

Oxidase. Reaction mixtures (1 ml final volume) contained xanthine
oxidase (0.05 U) and DMPO (60 mM) in 50 mM NaCl, pH 7.0. Additions
were as follows. (A) no xanthine; (B) plus xanthine (0. 33 mM), (C)

plus xanthine (o. 33 mM). plus ADP (1.25 mM); (D) plus xanthine (o. 33
mM), plus catalase (200 U).

 

74

 

 

IOG

 

75

Figure 7. ESR Spin Trapping of Hydroxyl Radical Generated by Xanthine
Oxidase and Rat Liver Ferritin. Reaction mixtures (1 ml final volume)
contained rat liver ferritin (3 mM Fe3+), xanthine (0.33 mM) and 60 mM
DMPO added at two minutes after initiation except where indicated in
50 mM NaCl, pH 7.0. Additions were as follows: (A) plus 0.10 U
xanthine oxidase with DMPO added at zero time; (B) plus 0.10 U
xanthine oxidase; (C) plus 0.10 U xanthine oxidase, plus 200 U
catalase; (D) plus 0.10 U xanthine oxidase, plus 1.25 mM ADP; (E) plus
0.05 U xanthine oxidase, plus 1.25 mM ADP, plus 200 U catalase.

76

lg5

DISCUSSION

The results of this study demonstrate that only 1-2% of the total
cytosolic iron from iron-loaded rats is present in a low molecular
weight form. The existence of such iron complexes in erythrocytes has
been documented but the demonstration of similar complexes in other
tissues, including liver, has not been achieved. In spite of the
largely hypothetical nature of such complexes they are generally
proposed to be a source of iron for the promotion of lipid
peroxidation lfl.l£19° The majority of the iron was recovered in
ferritin, in agreement with recent reports that demonstrate an
increased synthesis of ferritin in response to iron loading (nu).

As iron overload is known to result in enhanced lipid
peroxidation and toxicity it was of interest to assess the potential
for ferritin to serve as a source of iron for the formation of
complexes capable of initiating oxidative damage. Accordingly, it was
demonstrated that rat liver ferritin can provide the iron necessary to
support lipid peroxidation in a model system consisting of
phospholipid liposomes and xanthine oxidase. In the absence of either
ferritin or xanthine oxidase little MDA formation was observed
indicating that maximal activity required both xanthine oxidase and a
source of iron. It was found that rates of lipid peroxidation were
directly proportional to the amount of ferritin present. The data
depicted utilize ferritin containing 200 uM iron so that rates in the
absence of catalase are evident. Superoxide dismutase (100 U/ml)
inhibited peroxidation by 82 and 97% in the absence and presence of
catalase respectively, indicating that 027 was required for

77

78

peroxidation to occur. Other investigators had previously proposed
that xanthine oxidase was capable of releasing iron from ferritin
directly (26, 27), however this was before it was recognized that
xanthine oxidase produces 027 (32).

Several lines of evidence indicate that the primary function of
021 in promoting ferritin-dependent lipid peroxidation is the release
of iron from ferritin. The release of iron from ferritin was
continuously monitored spectrOphotometrically by complexing the iron
with 2,2'-dipyridyl. This chelator was chosen since it has high
affinity for iron, does not mobilize iron from ferritin in the absence
of a reductant, and forms a stable ferrous complex with a
characteristic absorbance maxima at 520 nm (2"). Incubation of
xanthine oxidase, xanthine, and ferritin with dipyridyl resulted in an
increase in A520 with time, indicative of ferrous iron release from
ferritin. Superoxide dismutase inhibited this iron release by 95%.
Potassium superoxide prepared in crown ether was similarly capable of
mobilizing iron from ferritin and $00 also inhibited this reaction.
Lastly, results of ESR spin trapping experiments also demonstrate an
interaction between ferritin and 02? generated by xanthine oxidase.
The DMPO-OOH adduct signal was observed in incubations containing
xanthine, xanthine oxidase and DMPO. However, the addition of
ferritin resulted in a significant diminuition of the EPR signal
intensity, suggesting that ferritin effectively competes with DMPO for
02-.

Superoxide-dependent initiation of lipid peroxidation is
generally thought to be via an iron-catalyzed Haber-Weiss mechanism.

The role of 027 in this process is two-fold, generation of H202 by

79

nonenzymatic dismutation (Reaction 2) and reduction of ferric iron to
ferrous (Reaction 1). However, it is believed that the concentration
of free or low molecular weight chelated forms of iron within cells is
miniscule, or perhaps nonexistent, therefore whether this mechanism is
operative in £129 remains speculative. In fact, one of the major
cellular protective mechanisms against oxidative cytolysis may be the
sequestration of redox active metals such as iron or copper in
unreactive states. In this context, this work suggests a novel role
for 027 in promoting toxicity, that is the reductive release of iron
from ferritin, thereby potentially increasing the low molecular weight
iron pool capable of undergoing redox reactions leading to the
formation of stronger oxidants.

The hydroxyl radical is the most widely proposed initiator of
lipid peroxidation and can be formed by Fenton's chemistry (Reaction
3). This would be a likely mechanism for initiation of lipid
peroxidation in the xanthine oxidase system as H202 is generated
directly by the enzyme (32) as well as by 027 dismutation and could
react with the ferrous iron released from ferritin. Hydroxyl radical-
dependent initiation of lipid peroxidation is often inferred from the
ability of catalase to inhibit lipid peroxidation by scavenging H202
(“5), thus preventing 'OH formation. However, the addition of
catalase to the xanthine oxidase and ferritin system used in this
study resulted in significant stimulation of rates of MDA formation.

Increasing xanthine oxidase activity, which would result in greater

H202 production, inhibited lipid peroxidation. This was in spite of
the greater rates of iron release from ferritin occurring at increased

xanthine oxidase activity. Even in the presence of catalase high

80

levels of xanthine oxidase activity still resulted in a decrease in
the rate of MDA formation, presumably because the generation of H202
exceeded the catalatic capacity of catalase. Although the amount of
H202 produced by xanthine oxidase is dependent upon factors such as
substrate concentrations (xanthine and oxygen) and pH (#6), the data
demonstrate that with 0.005 U/ml of xanthine oxidase activity no
further stimulation of lipid peroxidation was observed above 100 U/ml
of catalase.

Similarly, when catalase was added to another xanthine oxidase-
dependent system in which ferritin was replaced with ADP--Fe3+ (at iron
concentrations similar to that released from ferritin by xanthine
oxidase) enhancement of lipid peroxidation was also noted. These
results indicated that the effect of catalase was not unique to
ferritin per se (i.e., due to the facilitation of iron release from
ferritin) but rather that H202 is inhibitory to lipid peroxidation
when iron is supplied at low concentrations. This premise was
supported by the results of studies of iron release from ferritin.
The addition of catalase caused a further increase in dipyridyl-Fe2+
formation. Catalase may cause this apparent increase in iron release
by preventing H202-dependent oxidation of released ferrous iron or of
the dipyridyl'Fe2+ complex (H7) as dipyridyl-Fe3+ has little
absorbance at 520 nm. To distinguish between these possibilities
varying concentrations of dipyridyl—Fe2+ (0-100 nmol Fe2+lml) were
incubated with xanthine and xanthine oxidase (0.025 U/ml) or H202
(1mM). No oxidation of the ferrous complex was observed in either
case over a 30 minute period (results not shown). Therefore it

appears that the apparent increase in the amount of iron released from

81

attributable to its

atalase, is

rior to its
n of the

resence of c

in the p
errous iron p

ing oxidatio
d at saturating

ferritin,
f released f

prevention of oxidation 0

When dipyridyl was include

in apparent i

was

2* complex.
ron release

dipyridyl-Fe
her increase

concentrations no furt

ved at catal
e of the rele

n into ferrit

obser
The fat

incorporatio

As iron
in the prese

nce of H202,

may be that,
Alternatively,

it
and reincorporated into ferritin.

3 iron after
se possibilit

uled out by

ferrou
ies can be r

Neither of the
s not affect

emonstrated t

hat H20 2 doe

ze ferrous iron,

we have d
itin but appears to oxidi

tor of lipid

ase stimulate

ferr
peroxidation.
g that

indicatin
(DH, it is

of an initia
ion,

though catal
lipid peroxid

d MDA format
t dependent upon
ge all R202
ssibility

Al
ation is no
and some (OH

initiation of
le that catal

en generated.

gate this po

conceivab
peroxidation s

n the lipid ystem.
mannitol 3

still have be
{OH trap, w
ce or absenc

may
as included i

e of catalase,

mannitol, an
er the presen lightly

es of MDA for

In eith
mation, therefore initiation appears

idation was

ping eXperiments. The addition

sulted in th
al minutes af

e appearance

ter

when DMPO was added sever

82

initiation of the reaction to allow ferritin to interact with 027.
The intensity of the signal was found to correlate with ferritin
concentrations and therefore with the amount of iron released.
Importantly, the addition of catalase to the incubation resulted in a
marked decrease in DMPO-OH signal intensity, indicating that under
conditions in which ~0H formation is prevented, lipid peroxidation is
maximal.

The present studies demonstrate that ferritin can provide the
iron necessary for initiation of lipid peroxidation. Furthermore, the
data indicate that rapid oxidation of released ferrous iron produces
fOH but apparently precludes the formation of a more efficient
initiating species. However, irrespective of the mechanism, these
findings suggest that the proposed toxicity of 027 may be a result of
its ability to release iron from ferritin, with the released iron
functioning in the iron-oxygen redox reactions leading to the

initiation of lipid peroxidation.

10.

11.

12.

13.

LIST OF REFERENCES

J.M. McCord. Oxygen-derived free radicals in postischemic tissue
injury. New England J. of Medicine 312: 159-163 (1985).

B. A. Freeman, M. K. Topolsky, and J. D. Crapo. Hyperoxia
increases oxygen radical production in rat lung homogenates.
Arch. Biochem. Biophys. 216: R77- -48N (1982).

P. Hochstein. Futile redox cycling: Implications for
oxygen radical toxicity. Fund. Appl. Toxicol. 3: 215-217
(1983). ‘ ‘ ' ‘

B. Halliwell and J. M. C. Gutteridge. Oxygen toxicity, oxygen
radicals, transition metals and disease. Biochem. J. 219:
1-1" (198“) ' ' '

J.M.C. Gutteridge, G.J. Quinlan, and S. Wilkins. Mitomycin
c-induced deoxyribose degradation inhibited by superoxide
dismutase. FEBS Lett. 167: 37-“1 (198“).

I. Fridovich. Superoxide radical: An endogenous toxicant.
Ann. Rev. Pharmacol. Toxicol. 23: 239-257 (1983).

J.M. McCord and E.D. Day, Jr. Superoxide-dependent
production of hydroxyl radical catalyzed by the iron-EDTA
complex. FEBS Lett. 86: 139-1H2 (1978).

J.V. Bannister, W.H. Bannister, R.A.O. Hill, and P.J.
Thornalley. Enhanced production of hydroxyl radicals by the
xanthine-xanthine oxidase reaction in the presence of
lactoferrin. Biochim. Biophys. Acta 115: 116-120 (1982).

J.V. Bannister, P. Bellavite, A. Davioli, P.J. Thornalley,
and F. Rossi. The generation of hydroxyl radicals following
superoxide production by neutrophil oxidase. FEBS Lett.
150: 300- 302 (1982). '

C.C. Winterbourn. Hydroxyl radical production in body
fluids. Roles of metal ions, ascorbate, and superoxide.
Biochem. J. 198: 125-131 (1981).

J.M.C. Gutteridge, D.A. Rowley, and B. Halliwell.
Superoxide-dependent formation of hydroxyl radicals and
lipid peroxidation in the presence of iron salts. Biochem.
J. 296: 605-609 (1982).

A. Jacobs. Low molecular weight intracellular iron
transport compounds. Blood 59: u33~n39 (1977).

G.R. Bartlett. Phosphate compounds in rat erythrocytes and
reticulocytes. Biochem. Biophys. Res. Comm. 19: 1055-1062
(1976). ' ‘ ' ' ' '

83

14.

15.

16.

17.

18.

19.

20.

21.

22.

23.

2“.

25.

26.

84

S. Pollack and T. Campana. Low molecular weight non-heme
iron and a highly labeled heme pool in the reticulocyte.
Blood 56: 56N-566 (1980).

K.L. Fong, P.B. McCay, J.L. Poyer, H.P. Misra and 8.8. Keele.
Evidence for superoxide-dependent reduction of Fe3+ and its
role in enzyme-generated hydroxyl radical formation. Chem.
Biol. Inter. 15:77-89 (1976). '

C. G. D. Morley and A. Bezkorovainy. Identification of the
iron chelate in hepatocyte cytosol. IRCS Med. Sci. 11:
1106- -1107 (1983) ' "‘

P. M. Harrison. Ferritin: An iron-storage molecule. Sem.
Hematol 1n: 55-70 (1977).

P. Aisen and I. Listowsky. Iron transport and storage
proteins. In:‘ Annual Review of Biochemistry, Volume "9
(E.E. Snell, P.D. Boyer, A. Meister and C.C. Richardson,
eds.). PP. 357'393, Annual Reviews, Inc., Pala Alto, CA
(1980). ‘

 

R.R. Crichton. Iron uptake and utilization by mammalian
cells. II. Intracellular iron utilization. Trends
Biochem. Sci. 9: 283-285 (198“).

E.C. Theil. Ferritin: Structure, function, and regulation.
In: Iron Binding Proteins Without Cofactors or Sulfur
Clusters (E.C. Theil, G.L. Eichhorn, and L.G. Marzilli,

 

eds.). pp. 1-38, Elsevier, New York (1983).

S. Granick and L. Michaelis. Ferritin II. Apoferritin of
horse spleen. J. Biol. Chem. 1&7: 91-97 (19M3).

R.R. Crichton. Ferritin. Struct. Bond 11; 67-13“ (1973).

A. Mazur, S. Baez, and E. Shorr. The mechanism of iron
release from ferritin as related to its biological
prOperties. J. Biol. Chem. 213: 1N7-160 (1955).

R. R. Crichton, F. Roman, and F. Roland. Iron mobilization
from ferritin by chelating agents. J. Inorg. Biochem. 13:
305'316 (1980) ‘ ‘ "

J. Dognin and R.R. Crichton. Mobilization of iron from
ferritin fractions of defined iron content by biological
reductants. FEBS Lett. 33: 23u-236 (1975).

A. Mazur and A. Carleton. Hepatic xanthine oxidase and
ferritin iron in the developing rat. Blood 26: 317-322
(1965). ‘

27.

28.

29.

300

31.

32.

33.

3h.

35.

36.

37.

38.

39.

85

A. Mazur, S. Green, A. Saha and A. Carleton. Mechanism of
release of ferritin iron in vivo by xanthine oxidase. J.
Clin. Invest. 37: 1809- 1817 (1958).

R.W. T0pham, M.C. Walker, and M.P. Calisch. Liver xanthine
dehydrogenase and iron mobilization. Biochem. Biophys. Res.
Comm. 109: 12H0-12M6 (1982). ‘ ' '

J.M.C. Gutteridge, B. Halliwell, A. Treffry, P.M. Harrison,
and D. Blake. Effect of ferritin-containing fractions with
different iron loading on lipid peroxidation. Biochem. J.

£93: 557-560 (1983).

E.D. Wills. Mechanisms of lipid peroxide formation in
animal tissues. Biochem. J. 22: 667-676 (1966).

E.D. Wills. Lipid peroxide formation in microsomes. The
role of non-haem iron. Biochem. J. 113: 325-332 (1969).

J. M. McCord and I. Fridovich. The reduction of cytochrome c
by milk xanthine oxidase. J. Biol. Chem. 2H3: 5753-5760
(1968). '

J. M. McCord and I. Fridovich. Superoxide dismutase. An
enzymic function for erythrocuprein (hemocuprein). J. Biol.
Chem. 2AA: 60u9-6055 (1969). ’ '

R.F. Beers, Jr. and I.W. Sizer. A spectrophotometric method
for measuring the breakdown of hydrogen peroxide by catalase.
J. Biol. Chem. 195: 133-1N0 (1952).

J.W. Halliday. Immunoassay of ferritin in plasma. In:
Methods in Enzymology, Volume 8H (J. Langone and J. Van

 

Vunakis, eds.) pp. 1N8-171, Academic Press, New York (1982).

T.C. Pederson and S.D. Aust. Aminopyrine demethylase.
Kinetic evidence for multiple microsomal activities.
Biochem. Pharmacol. 12: 2221-2230 (1970).

J. Folch, M. Lees, and G.R. Sloane-Stanley. A simple method
for the isolation and purification of total lipids from
animal tissues. J. Biol. Chem. 226: H97-509 (1957).

T.C. Pederson, J.A. Buege and S.D. Aust. Microsomal
electron transport. The role of reduced nicotinamide
adenine dinucleotide phosphate-cytochrome c reductase in
liver microsomal lipid peroxidation. J. Biol. Chem. 238:
7139-71M1 (1973L ' ‘ ‘

G.R. Bartlett. Phosphorus assay in column chromatography.
J. Biol. Chem. 23“: H66-N68 (1959).

no.

“1.

“2.

"3.

H4.

"5.

H6.

”7.

N8.

86

J.A. Buege and S.D. Aust. Microsomal lipid peroxidation.
In: Methods in Enzymology, Volume 52 (S. Fleischer and L.
Packer, eds.) pp. 302-310, Academic Press, New York (1978).

P.B. Brumby and V. Massey. Determination of non-heme iron,
total iron, and cOpper. In: Methods in Enzymology, Volume
10 (R. w. Estabrook and M. D. Pullman, eds.) pp. u63-u7u,
Academic Press, New York (1967).

 

G.W. Ruddock, J.A. Raleigh, and G.L. Greenstock. Reactivity
of chemically generated superoxide radical anion with
peroxides as determined by competition kinetics. Biochem.
Biophys. Res. Comm. 192: SSH-560 (1981). ‘

G.R. Buettner. The spin trapping of superoxide and hydroxyl
radicals. In: Superoxide Dismutase, Volume II (L.W.
Oberley, ed.) Pp. 63-81, CRC Press, Boca Raton (1982).

G. E. Shull and E. C. Theil. Regulation of ferritin mRNA: A
possible gene-sparing phenomenon. J. Biol Chem. 258:
7921-7923 (1983) '

R.W. Kellogg, III and I. Fridovich. Superoxide, hydrogen
peroxide and singlet oxygen in lipid peroxidation by a
xanthine oxidase system. J. Biol. Chem. 250: 8812-8817
(1975). ' ' ‘ ’

I. Fridovich. Quantitative aspects of the production of
superoxide anion radical by milk xanthine oxidase. J. Biol.
Chem. 2N5: u053-1057 (1970). '

G.W. Winston, W. Harvey, L. Berl and A.I. Cederbaum. The
generation of hydroxyl and alkoxyl radicals from the inter-
action of ferrous bipyridyl with peroxides. Biochem. J.

_2_1_6: 1115-1421 (1983).

M. Wauters, A.M. Michelson, and R.R. Crichton. Studies on
the mechanism of ferritin formation; FEBS Lett. 91:
276-280 (1978). ' '

CHAPTER II

PARAQUAT AND FERRITIN-DEPENDENT LIPID PEROXIDATION

87

ABSTRACT

A lipid peroxidation system consisting of phospholipid liposomes,
paraquat, ADP, and NADPH-cytochrome PM50 reductase was constituted
using ferritin as the sole source of iron. Lipid peroxidation was
inhibited by SOD, essentially not affected by mannitol, but markedly
stimulated by catalase. Similar effects of these scavenging agents
were observed in incubations void of ADP. These data suggest that
02?, produced by the redox cycling of paraquat, can release iron from
ferritin and thereby promote lipid peroxidation. The effects of
catalase and mannitol suggest that the initiation of peroxidation, in
either the presence or absence of ADP, is not significantly dependent
upon the hydroxyl radical produced via an iron-catalyzed Haber-Weiss

reaction.

88

INTRODUCTION

Paraquat (methyl viologen, 1-1-dimethy1-N,U'-bipyridylium
dichloride), is a broad spectrum herbicide widely used as a
non-selective weed killer which has been shown to be highly lethal to
man and animals (1-A). The target organ appears to be the lung, most
probably because of active uptake (5) but biochemical changes are also
seen in kidney, thymus, and adrenals of paraquat-treated animals
(6-9). Damage is thought to be related to increased 02? production,
formed as a result of NADPH-cytochrome PHSO reductase catalyzed
reduction and autoxidation of the herbicide (10-17). As 02? is
relatively unreactive towards most organic compounds in aqueous
media (18), others have suggested NADPH or GSH depletion as mechanisms
of toxicity (19-20).

However, it must be considered that in the presence of transition
metal ions such as copper or iron, 02? is converted to a more
reactive, partially reduced oxygen species capable of initiating lipid
peroxidation. The nature of the initiating species is controversial
but is generally thought to be fOH or an iron-oxygen complex (21-25).
Irrespective of mechanism, there is a requirement for metal ions for
formation of an initiator. Thus, while it is likely that iron may
have an important role in mediating paraquat toxicity, the ability of
physiological iron to participate in such redox reactions is
relatively unknown. Most investigators envision an intracellular low
molecular weight iron chelate pool, however, the identity of this
complex(es) remains to be established. Nucleotide-chelated iron

89

9O

complexes have been isolated from erythrocytes (26,27), however, their
existence in other tissues has not been conclusively demonstrated.

The majority of iron in most cells is stored within ferritin as
ferric hydroxide complexed with phosphate (28).Mobilization of iron
from within ferritin appears to require reduction (29), but biological
reductants which release iron at rates considered to be of
physiological significance have not been identified (30). Previous
work in this laboratory demonstrated that 027, generated by xanthine
oxidase, can release iron from ferritin and promote the peroxidation
of phospholipid liposomes (31).

As paraquat also produces 027 by redox cycling, we have
investigated the ability of paraquat and NADPH-cytochrome P-HSO
reductase to initiate the peroxidation of liposomes using ferritin as
the sole source of iron. The results show that the redox cycling of
paraquat promotes ferritin-dependent lipid peroxidation. Inhibition
of peroxidation by SOD suggests that 027 is required for ferritin iron
release, as was observed with xanthine oxidase (31). These data
indicates that ferritin may function in 1119 as a source of iron for
the potentiation of peroxidative damage associated with paraquat

intoxication.

MATERIALS AND METHODS

Materials

Paraquat dichloride, NADPH (Type III), ADP (Grade III),
cytochrome 0 (Type VI), 2-TBA, butylated hydroxytoluene, mannitol and
u,7-dipheny1-1,10-phenanthroline were purchased from Sigma Chemical
Company (St. Louis, MO). Horse spleen ferritin was a product of
United States Biochemical Corporation (Cleveland, OH). EDTA and H202
were obtained from Mallinckrodt Chemical Company (Paris, KY) and
ferric chloride was purchased from Baker Chemical Company
(Phillipsburg, NJ). Sephadex G-25 was obtained from Pharmacia
(Piscataway, NJ) and Chelex 100 from BioRad Laboratories (Richmond,
CA). All other chemicals were of analytical grade or better and used
without further purification. All buffers and reagents were passed
through Chelex 100 ion-exchange resin to free them of contaminating
transition metal ions.
Enzymes

Bovine erythrocyte SOD (EC 1.15.1.1) was obtained from Sigma
Chemical Company and catalase (EC 1.11.1.6) was purchased from
Millipore (Freehold, NJ). Protease-solubilized NADPH-cytochrome PASO
reductase (EC 1.6.2.“) was prepared from liver microsomes of male
Sprague-Dawley rats (250-300 3), previously pretreated with 0.1%
phenobarbital in their drinking water for 10 days, as previously
described (32). The reductase was desalted on Sephadex G-25 columns

equilibrated with 0.3 M NaCl, pH 7.0 prior to use with a unit of

91

92

activity defined as 1 nmol cytochrome c reduced/min/ml. Superoxide
dismutase activity was measured by the method of McCord and
Fridovich (33) and catalase activity by the procedure of Beers and
Sizer (3A).
Preparation of Ferritin and Assay for Total Iron

Horse spleen ferritin was incubated on ice in 10 mM EDTA for 1
hour and passed over a Sephadex G-25 column equilibrated with 0.3 M
NaC1, pH 7.0 to remove loosely associated iron. Total iron was
determined essentially by the method of Brumby and Massey (35) with
minor modification (31).

Preparation of Microsomal Lipid and Liposomes

 

Rat liver microsomes were isolated by the procedure of Pederson
and Aust (36) and microsomal lipid was extracted by the method of
Folch 33 El- (37). All solvents utilized were purged with argon and
all steps performed at A°C to minimize autoxidation of unsaturated
lipids. PhOSpholipid liposomes were subsequently prepared by
indirect, anaerobic sonication (38) and lipid phosphate determined by
the method of Bartlett (39).

Lipid Peroxidation Assays

 

Paraquat-dependent peroxidation of liposomes was performed by
incubating liposomes (1 nmol lipid phosphate/ml) with NADPH,
NADPH-cytochrome PASO reductase, paraquat, ferritin and ADP as
specified in the figure legends. In some experiments lipid
peroxidation assays were performed without ADP. Reaction mixtures
were constituted in 0.25 M NaCl,pH 7.0, and incubated at 37°C in a
Dubnoff metabolic shaker under an air atmosphere. Although

unbuffered, incubations remained at pH 7.0 throughout the course of

93

the experiments. Reactions were initiated by the addition of NADPH
and peroxidation was monitored by taking aliquots from the incubations
at 0, 10, 20, and 30 min. and measuring MDA formation by the TBA assay
(A0).Butylated hydroxytoluene (0.03 volumes of 2% BHT in ethanol) was
added to the TBA reagent to prevent further peroxidation of lipid

during the assay procedure.

RESULTS

Paraquat and Ferritin-Dependent Lipid Peroxidation

As shown in Figure 1, incubation of NADPH-cytochrome PNSO
reductase, paraquat, and ferritin promoted peroxidation of
phospholipid liposomes as evidenced by MDA production. Increasing the
activity of NADPH-cytochrome PA50 reductase in the reaction mixture
beyond 0.025 U/ml had an inhibitory effect on MDA formation. The
addition of catalase (50 U/ml) markedly promoted the rate of lipid
peroxidation at all NADPH-cytochrome PNSO reductase activities used.
In addition, the activity of NADPH-cytochrome PASO reductase yielding
the greatest rate of lipid peroxidation shifted from 0.025 U/ml to
0.05 U/ml when catalase was added.

As shown in Figure 2A, increasing the concentration of paraquat
beyond 1.0 mM also had an inhibitory effect on rates of lipid
peroxidation. In the absence of paraquat, only very low rates of
peroxidation were observed even when ferritin was present in the
reaction mixtures. On the other hand, increasing the ferritin
concentration (in the presence of paraquat) resulted in a concomitant
increase in rates of lipid peroxidation (Figure 28).

Effect of SOD, Catalase, and Mannitol on Paraquat and Ferritin-

Dependent Lipid Peroxidation

 

Ten units/ml of SOD inhibited the paraquat and ferritin-dependent
lipid peroxidation (data not shown) indicating that initiation of
lipid peroxidation in this system was 027-dependent. As catalase
stimulated the rate of paraquat and ferritin-dependent lipid

peroxidation (Figure 1), the effect of increasing catalase
94

..mco:umz new mHmHLmumz

CH mm <9: Lou nommmmm one: mLprHs coHpommL on» Bonn muoscqa< .o.~ ma .Howz z mm.o ad
mmmposump omzm maonnoou>o1mmo<z no mwfiua>fipom mcfizgm> new A28 N.ov mmo<z .Aumumofiucfi econ:
Hs\= omv emmamsmo .Azs m.ov passages .Azs m.mv age .A+mmm :5 _v erHLLmL .AHa\mpmnamona
nHaHH Hoe: FV mmeomoqfia ufiafiaonamona omchucoo Amssao> Hmcwu as m.mv mogsuxas coguommm
.mmmamumo no mocomn< Lo monommpm on» :H mnfiqfiaonamonm uo :oHumonoLom ucmucoaoorchfignom
vcm pmscmpmm co >uH>Hpo< mmmuosumm omam waoncoouzo1mmo<z mcfimgm> no uommmm one .P onsmfim

96

 

253 $28an
No :0 mod .08

' . ' . 1)

$2200 ..

3238 +

_.O

«3
o

IUJ/Ulw/ VOW I014m

m‘
d

 

ho

 

97

. .muosum: cam mHmHLmumz

:H umumofivcH mm ommmmmm mm: uzmucoo <oz .zs m.o mm: umscmuma Amv CH cam + mm :8 —

mm: :HuHLLmu A<V :H .o.~ ma .Homz z mN.o 2H Asta mmo.ov mmmposomg omzm oBOLSOOp 01mma¢z
new Azs m.ov maoaz .Aze m.mv age .Aasxmsmeamone UHQHH Hos: PS meaomoaaa uHaHHogemona
omcfimucoo Awasao> Hmcfiu Ha m.mv mouspxfls coHpommm .mufiafiaonamonm mo coHpmcongmm

co :oHumLucmocoo Amv cfiufignom Lo A<v pmsumgmm mcfizgm> mo uoouum one .m ogsmfim

98

 

 

    

 

 

 

 

 

i~. £2 e -'.-

O O O O
le°ugwlvow low u

f».

O

le°U!w/vow10wu

 

1.0 2.0 3.0

0.1

 

Ferritin(umol Fe/ml)

PQ(mM)

99

concentrations was assessed. As shown in Figure 3, 50 U/ml of
catalase gave maximum rates of lipid peroxidation with 0.025 U/ml of
NADPH-cytochrome PN50 reductase activity. The marked stimulation of
peroxidation by catalase suggests that fOH is not involved in the
peroxidative process. These results were supported by the inability
of mannitol, in the absence of catalase, to inhibit peroxidation up to
50 mM mannitol (Figure A).
Paraquat and Ferritin-Dependent Lipid Peroxidation in the Absence
of ADP .

The effect of varying ADP concentration on paraquat and
ferritin-dependent lipid peroxidation is shown in Figure 5.
Increasing the ADP concentration resulted in a decrease in the rates
of lipid peroxidation with the greatest rate of lipid peroxidation
obtained in the absence of ADP. Others have suggested that the
chelation of iron with ADP may affect the mechanism of initiation of
lipid peroxidation (N1). Therefore, the effects of SOD, catalase and
mannitol on paraquat and ferritin-dependent lipid peroxidation in the
absence of ADP were also investigated. Superoxide dismutase again
inhibited the paraquat and ferritin-dependent lipid peroxidation at 10
units/ml, (data not shown) indicating again that initiation of lipid
peroxidation in this system is 027-dependent. Catalase also
stimulated the rate of lipid peroxidation but to a lesser extent than
was observed in the same system with ADP (Figure 6).

The effect of varying mannitol concentration on paraquat and
ferritin-dependent lipid peroxidation in the absence of ADP, either
in the presence or absence of catalase, is shown in Figure 7. As

shown previously (Figure 6) catalase stimulated activity slightly.

lOO

.Aze m. ov nonconma

m.ov moo<z .Aas\opmnamona UHQHH Hoe:
as m.mv mogsuxfis cofiuooom

.ooocpo: new maofipoumz ca nonagoooo mm nonopfiucosc no:
«a: .o.e me .Homz 2 mm. o as ommflmpmo Lo soaps>spem messgm> new “.mom :3 _v eflsflegou

.Azs m. NV ma< .AHa\= mmo. ov oomposeog omen osoesoopzo maaaz

ufizs

FV moaoooafla nfiqfiaonaooca cocfimucoo Aossao> Hogan
.mo< mo oocomogm on» :H :Hpfipnom new umsoopom an nonmampmo

mquHHonqmonm no cofiuocaxogom co >ufi>auo< oomaopmo mcamgo> no uooumm one .m ogsmam

101

 

 

 

A A A

 

' 1'0 25

 

 

<1; 10 N

o 0' o'
lwl'ugw/vow |ou1 u

o'

IOO

50

CotoloseiU/ml)

102

.nnonnoo «a: non noawoww ono: oonsuxas noHnowon on» song wuo:c«a< .0.» ma .Huwz z mm.o

nH Hogannwa no mnofinwnnnoonoo mnwznw> nnw Aas\n mmo.ov owwposnon,om=m osonnoon501nmo¢z
.Aza m.ov mwaaz .Azs m.ov swsewnwe .A+mon :8 FV enunnnon .Azs m.mv nae .nas\oswnamona
nHQHH Hoe: FV mosomoaaa nHQaHonqwona nonHwnnoo Aoasao> Hwnau as m.mv wonsnxfis

noHnowom .omwawnwo no oonomnn nnw mo< no oonowonm onu n“ wanHHonamonm no noHuwononom
nnonnoaoo1nfinwnnom nnw nwncwnwm no nofiuwnnnoonoo Honfinnw: mnH>Lw> no noomum one .3 onsmflm

103

 

 

0.3

 

A A

N .—

0 O

IWI'UW/VOW Iowu

 

20 30 4o 56
Monnitol(mM)

IO

104

Figure 5. The Effect of Varying ADP Concentration on Paraquat and
Ferritin-Dependent Peroxidation of Phospholipids. Reaction mixtures
(2.5 m1 final volume) contained phospholipid liposomes (1 nmol lipid
phosphate/ml), ferritin (1 mM Fe +), paraquat (0.5 mM), NADPH (0.2
mM), NADPH-cytochrome PASO reductase (0. 025 U/ml) and varying
concentrations of ADP in 0. 25 M NaCl, pH 7. 0. MDA content was
measured as in Materials and Methods.

105

 

 

 

0.6

5.0

 

 

4. . 2
O O

.E\.:_E\<o 2 .08 c

2.0 3.0 4.0
AD P (mM)

I.O

106

Figure 6. The Effect of Varying Catalase Activity on Paraquat and
Ferritin-Dependent Peroxidation of Phospholipids in the Absence of ADP.
Reaction mixtures (2.5 m1 final volume) contained phospholipid '
liposomes (1 nmol lipid phosphate/ml), ferritin (1 mM Fe3*), paraquat
(0.5 mM), NADPH (0.2 mM), NADPH-cytochrome PASO reductase (0.025

U/ml), and varying activities of catalase in 0.25 M NaCl, pH 7.0.

Aliquots of incubations were assayed for MDA as outlined in Materials
and Methods.

11 mol MDA/min./ml

107

 

 

 

2'5 50
C atalaseW/ml)

.1

 

J—u—e

100

108

Figure 7. The Effect of Varying Mannitol Concentration on Paraquat
and Ferritin-Dependent Peroxidation of Phospholipids in the Presence
or Absence of Catalase. Reaction mixtures (2.5 m1 final volume)
contained phospholipid liposomes (1 nmol lipid phosphate/ml), ferritin
(1 mM Fe3+), paraquat (0.5 mM), catalase (10 U/ml, where indicated),
NADPH (0.2 mM), NADPH-cytochrome PMSO reductase (0.025 U/ml) and
varying concentrations of mannitol in 0.25 M NaCl, pH 7.0. Aliquots
of reaction mixtures were assayed for MDA content as described in
Materials and Methods.

11 mo! M DA/min./ml

109

 

 

 

 

 

’1
+Cotolase
0.6 1 1
05‘ T
-Cotalase
~O
0A:
1 1
O ’E J - 4:
I

5 I0 270 30 40 50
Mannitol (m M)

110

The effects of mannitol were somewhat variable at low concentration
but it was essentially without effect in both the presence and absence

of catalase.

DISCUSSION

These studies demonstrate that the redox cycling of paraquat can
promote lipid peroxidation in a model system consisting of
phospholipid liposomes, NADPH-cytochrome PASO reductase, and ferritin.
In the absence of either ferritin, NADPH-cytochrome PNSO reductase, or
paraquat almost no MDA formation was observed, demonstrating a
requirement for both 027 and a source of iron. Previous work using a
xanthine oxidase system had shown that the primary function of 027 in
initiating ferritin-dependent lipid peroxidation is the release of
ferrous iron from ferritin (31). The inability of NADPH-cytochrome
9150 reductase to directly reduce ADP-Fe3* (32), and the inhibition of
peroxidation by $00, provides further evidence that 027 produced by
reduced paraquat autoxidation releases ferritin bound iron.

As iron is released from ferritin in the ferrous state (31), and
H202 is formed by the non-enzymatic dismutation of 027, the generation
of fOH by Fenton's reaction in this system seemed likely. Thus the
effect of mannitol, an fOH trap, on liposomal peroxidation was
assessed. A wide range of mannitol concentrations up to 50 mM had no
effect on rates of MDA formation. Therefore, initiation of
peroxidation in this system appeared to be independent offOH.

To examine this possibility further, catalase was added to
scavenge H202 and prevent -0H formation. Interestingly, catalase
markedly stimulated paraquat and ferritin-dependent lipid peroxidation.
Increasing either NADPH-cytochrome PASO reductase activity or paraquat

concentration, both of which increase 02? production, inhibited

111

112

peroxidation presumably by increasing H202 production. Even in the
presence of catalase, high NADPH-cytochrome PASO reductase activity
still resulted in a decrease in the rate of MDA formation, presumably
as the catalatic activity of catalase was exceeded. Accordingly, at
NADPH-cytochrome P050 reductase activities greater than 0.05 U/ml
increasing the concentration of catalase resulted in a further
stimulation of lipid peroxidation (results not shown). These data
demonstrate that with 0.025 U/ml of NADPH-cytochrome PHSO reductase
activity and 0.5 mM paraquat no further stimulation of peroxidation
was observed beyond 50 U/ml of catalase, suggesting that all H202 was
effectively scavenged.

Stimulation of liposomal peroxidation by catalase was previously
reported using ferritin and xanthine oxidase (31). A negative
correlation between fOH production and lipid peroxidation was
demonstrated with ESR spin trapping. Iron release studies suggested
that H202 did not affect iron release from ferritin but inhibited
peroxidation by facilitating oxidation of the released ferrous iron.
It is likely that in the present system the H202, generated by
dismutation of 027 or perhaps directly by paraquat radical
autoxidation (H2), similarly functions to prevent the more efficient
formation of an initiator other than -0H.

The greatest rate of MDA formation was observed in the absence of
ADP. Thus, it was of interest to determine whether ADP affected the
mechanism of initiation of peroxidation. Therefore, the effects of
SOD, catalase and mannitol on lipid peroxidation in the absence of ADP
were also examined. As obtained in the presence of ADP, SOD

significantly inhibited peroxidation. Therefore, 02? remains an

113

absolute requirement for peroxidation, even in the absence of ADP.
Catalase again stimulated rates of peroxidation, however the degree of
stimulation was less than in the presence of ADP. This effect
suggests that ADP affects the rate of 027 dismutation, or that
oxidation of free ferrous is slower than that of ADP-Fe2+ as has been
shown (”3), thus peroxidation would not be as markedly enhanced by
catalase. Since catalase still stimulated MDA formation in the
absence of ADP, initiation of lipid peroxidation still appeared to be
independent of :OH. Contribution of fOH to initiation of lipid
peroxidation in the absence of chelators (EDTA or ADP) was recognized
by Tien £3 31. (A1) using a Fenton-type reaction. In the present
system, the inability of mannitol to inhibit lipid peroxidation
indicated that the fOH may not be involved. Thus, it appears that the
addition of ADP does not significantly affect the apparent
mechanism(s) of initiation of lipid peroxidation.

The toxicity of paraquat has been attributed to its enhancement
of cellular 027 production with concomitant increased lipid
peroxidation (10,13-17). Superoxide is relatively unreactive towards
most organic compounds (18) in aqueous media, thus it likely
participates in the formation of other more reactive species capable
of initiating peroxidation. Superoxide is generally proposed to
function in the iron-catalyzed Haber-Weiss reaction to generate fOH;
however, the indiscriminate reactivity of -0H and the potentially
limited availability of free iron has led to suggestions that fOH may

not be the initiator in vivo. This hypothesis is supported by the

 

inability of fOH scavengers to inhibit in vitro microsomal

peroxidation (an) or to ameliorate toxicity when given to paraquat

114

treated animals (A5). In most instances, SOD effectively inhibits
paraquat mediated lipid peroxidation thus its participation must be
via a mechanism other than the Haber-Weiss reaction. The above
findings can be rationalized by these studies which suggest that
paraquat may enhance lipid peroxidation by generating 02? which can
release iron from ferritin. Thus, what normally provides a secure
means of sequestering iron (ferritin) now may serve as a source of
iron for the initiation of peroxidation.

While these studies demonstrate that paraquat can stimulate NADPH
and ferritin-dependent lipid peroxidation, the role of lipid
peroxidation in the in 1112 toxicity of paraquat remains to be
determined. Certainly, lipid peroxidation is not the only potentially
destructive reaction and may not wholly account for the observed
pulmonary toxicity. We have shown, by Ouchterlony double diffusion
analysis, that rat lung microsomes contain significant quantities of
ferritin (C.E. Thomas and S.D. Aust, unpublished). Therefore,
interaction of 027 with ferritin in the lung may provide the iron

necessary for the initiation of lipid peroxidation.

10.

11.

12.

13.

LIST OF REFERENCES

C.M. Bullivant. Accidental poisoning by paraquat: Report of two
cases in man. Br. Med. J. 1: 1272-1273 (1966).

D.F. Kelly, D.G. Morgan, P.G.G. Darke, H. Pearson, and B.M.G.
Weaver. Pathology of acute respiratory distress in the dog
associated with paraquat poisoning. J. Comp. Pathol. 88:
275-29“ (1978). ‘

R. D. Kimbrough and T. B. Gaines. Toxicity of paraquat to rats
and its effect on rat lungs. Toxicol. Appl. Pharmacol. 11:
679-690 (1970). ' ' "

R. E. Murray and J. E. Gibson. A comparative study of paraquat
intoxication in rats, guinea pigs and monkeys. Exp. Mol. Pathol.
17: 317-325 (1972) ‘ ' '

M. S. Rose, L. L. Smith and J. Wyatt. Evidence for energy-
dependent accumulation of paraquat into rat lung. Nature
(London) 252: 319-315 (1979).

D.G. Clark, T.F. McElligott and E.W. Hurst. The toxicity of
paraquat. Br. J. Ind. Med. 23: 126-132 (1966).

D.M. Conning, K. Fletcher and A.A.B. Swan. Paraquat and related
bipyridyls. Br. Med. Bull. g5: 2A5-299 (1969).

0.8. Vijeyaratnam and B. Corrin. Experimental paraquat
poisoning: A histological and electron-optical study of the
changes in the lung. J. Pathol. 103: 123-129 (1971).

0
A.P. Autor. Reduction of paraquat toxicity by superoxide
dismutase. Life Sci. 11: 1309-1319 (197”).

J.S. Bus, S.D. Aust and J.B. Gibson. Superoxide- and singlet
oxygen-catalyzed lipid peroxidation as a possible mechanism for
paraquat (methyl viologen) toxicity. Biochem. BiOphys. Res.
Commun. 58: 7u9-755 (1979). ' '

R.E. Brooks. Ultrastructure of lung lesions produced by ingested
chemicals. '1. Effect of the herbicide paraquat on mouse lung.
Lab. Invest. 25: 536-5A5 (1971).

H. M. Hassan and I. Fridovich. Superoxide radical and the oxygen
enhancement of the toxicity of paraquat in Escherischia coli. J.
Biol. Chem. 253: 81u3-81A8 (1978).

 

K. F. Ilett, B. Stripp, K. H. Menard, W. D. Reid and J. R. Gillette.
Studies on the mechanism of the lung toxicity of paraquat:
Comparison of tissue distribution and some biochemical parameters
in rats and rabbits. Toxicol. Appl. Pharmacol. 28: 216-226
(197A). ‘ ‘ ' '

115

1H.

15.

16.

17.

18.

19.

20.

21.

22.

23.

2H.

25.

116

J. S. Bus, S. 2. Cagen, M. Olggard and J.E. Gibson. A mechanism of
paraquat toxicity in mice and rats. Toxicol. Appl. Pharmacol.
35: 501-513 (1976). ' ‘

J.S. Bus, S.D. Aust and J.B. Gibson. Lipid peroxidation: A
possible mechanism for paraquat toxicity. Res. Commun. Chem.
Pathol. Pharmacol. 11: 31-38 (1975), - . A

M.A. Trush, E.G. Mimnaugh, E. Ginsburg and T.E. Gram. In vitro
stimulation by paraquat of reactive oxygen-mediated lipid

peroxidation in rat lung microsomes. Toxicol. Appl. Pharmacol.
60: 279-286 (1981) ' '

H.K. Fisher. Importance of oxygen and of pulmonary alveolar
surfactant in lung injury by paraquat. In: Biochemical
Mechanisms of Paraquat Toxicity_(A. P. Autor. ed.) pp. 57-65.
Academic Press, New York (1977).

 

I. Fridovich. Superoxide radical: An endogenous toxicant. Ann.
Rev. Pharmacol. Toxicol. 23: 239-257 (1983). '

L.L. Smith, M.S. Rose and I. Wyatt. The pathology and
biochemistry of paraquat. In: Oxygen Free Radicals in Tissue

 

Damage (Ciba Foundation Symposium 65) pp- 321-341, Excerpta

Medica, New York (1979).

H. Sies. R. Brigelius, H. Wefers, A. Muller and E. Cadenas.
Cellular redox changes and response to drugs and toxic agents.
Fund. Appl. Toxicol. 3: 200-208 (1983)

B. Halliwell and J. M. C. Gutteridge. Oxygen toxicity, oxygen
radicals, transition metals and disease. Biochem. J. 219: 1-1H
(198”). ‘ ‘ ‘ ‘ '

S. E. Fridovich and N. A. Porter. Oxidation of arachidonic acid in
micelles by superoxide and hydrogen peroxide. J. Biol. Chem.
256. 260-265 (1981). ‘ '

J.S. Bus and J.E. Gibson. Lipid peroxidation and its role in
toxicology. In:' Reviews in Biochemical Toxicology, Volume 1 (E.
Hodgson, J. R. Bend and R. M. Philpot, eds. ) pp. 125-1N9, ElseVier,
New York (1979)

 

S.D. Aust and B.A. Svingen. The role of iron in enzymatic lipid
peroxidation. In: Free Radicals in Biology, Volume 5 (W.A.
Pryor, ed.) pp. 1-28, Academic Press, New York (1982).

 

J.M.C. Gutteridge, D.A. Rowley and B. Halliwell. Superoxide-
dependent formation of hydroxyl radicals and lipid peroxidation
in the presence of iron salts. Detection of catalytic iron and
antioxidant activity in extracellular fluids. Biochem. J. 296:
605-609 (1982). ‘ '

26.

270
28.
29.
30.
31.

32.

33.

3M.

35.

36.

370

38.

117

G.R. Bartlett. Phosphate compounds in rat erythrocytes and
reticuloytes. Biochem. Biophys. Res. Commun. 19: 1055'1062
(1976). - ~ - - - -

G.R. Bartlett. Iron nucleotide in human and rat red cells.
Biochem. Biophys. Res. Commun. 19: 1063~1070 (1976).

P.M. Harrison. Ferritin: An ironvstorage molecule. Sem.
Hematol. 15‘ 55'70 (1977). ‘

E.C. Theil. Ferritin: Structure, function and regulation. In:
Iron Binding Proteins without Cofactors or Sulfur Clusters.
Volume 5 (E. c. Theil, G. L. Eichhorn and L. G. Marzilli, eds.) pp.

1-38, Elsevier, New York (1983).

A. Mazur, S. Baez and E. Shorr. The mechanism of iron release
from ferritin as related to its biological properties. J. Biol.
Chem. 213.1u7vv16o (1955). ‘ '

C. E. Thomas, L. A. Morehouse and S. D. Aust. Ferritin and
SUperoxidevdependent lipid peroxidation. 'J. Biol. Chem. 260:
3275'3280 (1985) ' '

L.A. Morehouse, C.E. Thomas and S.D. Aust. Superoxide generation
by NADPH—cytochrome PNSO reductase: ' The effect of iron chelators
and the role of superoxide in microsomal lipid peroxidation.
Arch. Biochem. Biophys. 232. 366-377 (198A).

J. M. McCord and I. Fridovich. Superoxide dismutase. An enzymic
function for erythrocuprein (hemocuprein). J. Biol. Chem. 2““:
60N9'6055 (1969). - . . -—-

R. F. Beers, Jr. and I. W. Sizer. A Spectrophotometric method for
measuring the breakdown of hydrogen peroxide by catalase. J.
Biol. Chem. 195: 133 1H0 (1952).

P.B. Brumby and V. Massey. Determination of nonheme iron, total
iron, and copper.‘ Methods Enzymol. 19‘ neg-u7u (1967).

T. C. Pederson and S. D. Aust. Aminopyrine demethylase. Kinetic
evidence for multiple microsomal activities. Biochem. Pharmacol.
19: 2221*2230 (1970). ‘

J. Folch, M. Lees and G.H. Sloane~Stanley. A simple method for
the isolation and purification of total lipids from animal
tissues. J. Biol. Chem. 226: ”97'509 (1957).

T.C. Pederson, J.A. Buege and S.D. Aust. Microsomal electron
transport. The role of reduced nicotinamide adenine dinucleotide
phosphate'cytochrome 0 reductase in liver microsomal lipid
peroxidation. J. Biol. Chem. gag: 713h-71fl1 (1973).

39.
M0.

111.

N2.

“3.

1111.

N5.

118

G. R. Bartlett. Phorphorus assay in column chromatography. J.
Biol. Chem. 23a: M66 A68 (1959). '

J. A. Buege and S. D. Aust. Microsomal lipid peroxidation.
Methods Enzymol. 52: 302-310 (1977).

M. Tien, B.A. Svingen and S.D. Aust. An investigation into the
role of hydroxyl radical in xanthine oxidasevdependent lipid
peroxidation. Arch. Biochem. Biophys. 216: 1u2—151 (1982).

A. Ledwith. Electron transfer reactions of paraquat. In:
Biochemical Mechanism of Paraquat Toxicity (A. P. Autor, ed.) pp.
21-38, Academic Press, New York (1977).

 

J.R. Bucher, M. Tien, L.A. Morehouse and S.D. Aust. Influence of
superoxide dismutase and catalase on strong oxidant formation
during autoxidation of ferrous chelates. In: Oxy Radicals and
Their Scavenger SystemSLAyolume I: Molecular Aspects (G. Cohen

 

 

and R.A. Greenwald, eds.) PP. 292'295, Elsevier Science
Publishing Co., Inc., New York (1983).

H. P. Misra and L. D. Gorsky. Paraquat and NADPHvdependent lipid
peroxidation in lung microsomes. J. Biol. Chem. 256: 999fl'9998
(1981). ' ' '

R.D. Fairshter, N.D. Vaziri, L.C. Dearden, K. Malott and M.
Caserio. Effect of dimethylthiourea on paraquat toxicity in rats.
Toxicol. Appl. Pharmacol. 13: 150*15" (198”).

CHAPTER III

REDUCTIVE RELEASE OF IRON FROM FERRITIN BY CATION
FREE RADICALS OF PARAQUAT AND OTHER BIPYRIDYLS

119

ABSTRACT

NADPchytochrome Ph50 reductase catalyzed reduction of paraquat
promoted the release of iron from ferritin. Aerobically, iron release
was inhibited approximately 60% by SOD while xanthine oxidase-
dependent iron release was inhibited nearly 100%. This suggests that
both 027 and the paraquat cation radical can catalyze the release of
iron from ferritin. Accordingly, under anaerobic conditions, the
paraquat radical mediated a very rapid, complete release of iron from
ferritin. Similarly, the cation free radicals of the closely related
chemicals, diquat and benzyl viologen, also promoted iron release.

ESR studies demonstrated that electron transfer from the paraquat
cation radical to ferritin accounts for the reductive release of iron.
The ferritin structure was not altered by exposure to the paraquat
radical and also retained its ability to reincorporate iron. These
studies indicate that release of iron from ferritin may be a common

feature contributing to free radical mediated toxicities.

120

INTRODUCTION

The toxicities associated with numerous drugs and chemicals is
often attributed to their ability to be metabolized to free radical
species. The free radicals of many of these compounds, such as
paraquat and related bipyridyls,undergo cyclic reduction and
autoxidation resulting in the production of partially reduced forms of
dioxygen such as 027 and H202 (1). Many studies have linked excessive
027 and/or H202 production with oxidative damage to numerous
biological macromolecules including lipids, proteins, and DNA (2).
However, neither 02? nor H202 is considered to be of sufficient
reactivity to directly initiate and/or promote oxidation of cellular
constituents (3). Therefore oxidants capable of initiating lipid
peroxidation and other deleterious oxidative processes are generally
prOposed to result from reactions between these species and transition
metals (A).

Although the initiating species has not been clearly identified,
numerous studies indicate that iron can be intimately involved in the
peroxidative process. Thus, the in XEXQ formation of an initiator is
likely to be contingent upon the availability of physiological iron.
Although iron metabolism has been the subject of considerable study
the possibility that iron may participate in these deleterious
oxidations in 1119 remains largely unexplored. At present the most
likely catalyst for mediating oxidative processes is thought to be a
low molecular weight intermediate functioning to transport iron
between transferrin and ferritin or between ferritin and other

ironvrequiring biochemical processes (5). Some evidence for such
121

122

complex(es) in erythrocytes has been presented (6,7) although in other
tissues such as liver a variety of potential chelators such as ADP (8)
or citrate (9) have been proposed but not conclusively identified.

While the existence of low molecular weight iron chelates remains
speculative, it is clear that the vast majority of iron, other than
heme iron as in hemoglobin, is present in the high molecular weight
proteins transferrin, ferritin, and hemosiderin (10). The relative
insolubility of ferric iron and its potential toxicity necessitates
its complexation by these proteins so it is likely that release of
iron from the protein(s) is necessary for catalysis of the
aforementioned reactions. Most of the intracellular iron is stored
within ferritin, a large spherical protein consisting of 2" subunits
arranged symmetrically around a hollow core in which ferric iron is
held in crystalline aggregates (11). Access to the central core is
via six narrow, hydrOphobic channels through which the iron can be
shuttled.

The release of ferric iron from ferritin appears to require
reduction (12). however positive identification of the physiological
reductant(s) has not been achieved. Harrison and colleagues (13)
suggest that reductants pass through the channels and interact
directly with the ferric hydroxide core, while others envison the
existence of oxidOvreduction sites on the interior of the channels
(1”). It has been demonstrated that reduced flavins, which are among
the most potent reductants for releasing ferritin iron, must at least
partially traverse the channels in order to release iron (15,16).
Furthermore, reductive mobilization of iron from ferritin appears to

be dependent upon formation of a complex at the surface of the iron

123

hydroxide core by the reductant (16). Thus, biological agents capable
of releasing iron from ferritin may be limited by their size as well
as their potential to form complexes. Another factor that must be
taken into consideration is the redox potential of the reductant as
recent work reports the reduction potential for ferritin iron to be
approximately ~230 mV at pH 7.0 (17).

We have previously demonstrated that 02?, generated by xanthine
oxidase, can reductively release iron from ferritin and promote the
peroxidation of phospholipid liposomes (18). Superoxide-dependent
iron release from ferritin by stimulated polymorphonuclear leukocytes
has also been demonstrated (19). Subsequent studies demonstrated that
the redox cycling of paraquat promoted ferritinvdependent lipid
peroxidation (20). These results support the reports which indicate
that paraquatvinduced lipid peroxidation is a result of enhanced

cellular 027 production (21,22). However, exposure of Escherichia

 

2211 to hyperbaric oxygen had little effect on paraquat induced
lethality (23) and toxicity has even been shown to be enhanced by low
oxygen concentrations (2"). There also exists conflicting data on the
ability of SOD to prevent or protect against paraquat toxicity
(25,26).

These apparent anomalous results suggest that the toxicity of
these compounds may not be just a function of redox cycling alone, or
of the free radical itself, but rather a combination of the two. The
paraquat cation radical has been reported to reduce ferric iron
(27,28), which, in conjunction with its relatively small size,
suggests it to be an excellent candidate for mediating iron release

from ferritin. Therefore, we have investigated the ability of

124

paraquat, and other chemicals capable of undergoing one electron redox
cycling, to reductively release iron from ferritin. In this paper we
demonstrate that several cation free radicals rapidly release
virtually all of the iron from within ferritin under anaerobic
conditions. Aerobically, iron release is mediated by both 027 and the
paraquat free radical. These studies suggest that radicalvdependent
release of iron from ferritin may contribute to the free radical
mediated toxicities of several chemicals by providing the iron

necessary for the formation of potentially harmful oxidants.

MATERIALS AND METHODS

Materials

Paraquat dichloride, benzyl viologen, NADPH (Type III),
cytochrome c (Type IV), bathophenanthroline sulfonate,
"-7vdiphenylv1,10vphenanthroline, Sephadex 0'25, and xanthine were
purchased from Sigma Chemical Company (St. Louis, MO). Glucose and
ferric chloride were obtained from Baker Chemical Company
(Phillipsburg, NJ) and thioglycolate from Fisher Chemical Company
(Fairlawn, NJ). Ultragel AcA AA was purchased from LKB (Bromma,
Sweden) and Chelex 100 from BiOvRad Laboratories (Richmond, CA).
Diquat was a generous gift from Dr. Lewis Smith of Imperial Chemical
Industries Ltd. (Macclesfield, Cheshire, United Kingdom) and Imferon
was a gift from Merrell Dow Pharmaceuticals (Cincinnati, OH).
Enzymes

Glucose oxidase (EC 1.1.3.“), xanthine oxidase (EC 1.2.3.2), and
bovine erythrocyte SOD (EC 1.15.1.1) were obtained from Sigma and
catalase (EC 1.11.1.6) was purchased from Millipore (Freehold, NJ).
The glucose oxidase contained significant adventitious iron (2.7 nmol
Fe3+/Unit of activity), therefore it was incubated with 10 mM EDTA for
one hour on ice and chromatographed over a Sephadex G-25 column
equilibrated with 0.3 M NaCl, pH 7.0 to remove contaminating iron.
The catalase was also chromatographed on Sephadex 0'25 and was assayed
as per Beers and Sizer (29) and xanthine oxidase was similarly treated
to remove ammonium sulfate and activity measured by aerobic reduction
of cytochrome c (30). Superoxide dismutase activity was measured by
the method of McCord and Fridovich (30). Protease solubilized

NADPH'cytochrome PHSO reductase (EC 1.6.2.u) was purified from
125 '

126

microsomes isolated from the livers of male SpraguevDawley rats
(250-300 g), previously pretreated with 0.11 phenobarbital

in their drinking water for 10 days, as previously described (31) with
the modification that enzyme eluting from the Sephadex 0,100 column
was subjected to affinity chromatography on 2',5"ADPvSepharose (32)
using buffers containing no detergent. The final preparation of
NADPchytochrome PASO reductase was chromatographed on a Sephadex 3-25
column with 0.3 M NaCl, pH 7.0 just prior to use, assayed, and
activity defined as 1 nmol cytochrome c reduced/min/ml.

Preparation of Ferritin and Iron Release Assays

Ferritin was purified from the livers of either control rats or
those given an intraperitoneal injection of Imferon (ironvdextran) as
described by Halliday (33) with slight modification (18).
Additionally, Ultragel AcA nu was used in place of Sepharose 6B and,
following purification, the ferritin was incubated on ice in 10 mM
EDTA for 1 hr and chromatographed on Sephadex 6'25 to remove traces of
adventitious iron. Total iron was determined essentially by the
bathophenanthroline method of Brumby and Massey (3A) with minor
modification (18).

Iron release from ferritin under aerobic conditions was
determined according to Thomas gt 2l° (18) with modification. The
ferrous iron chelator used was bathophenanthroline sulfonate (E -
22.1“ mM’1 cm’?) (35) and the increase in absorbance at 530 nm was
continuously monitored in a Cary 219 dual beam recording
spectrophotometer. For anaerobic experiments all solutions were
purged with argon, and glucose and glucose oxidase were added to the

incubation mixtures to scavenge any remaining oxygen. To prevent

127

ferrous oxidation by H202, catalase was also included. Incubation
mixtures (1 ml final volume) were constituted in 50 mM NaCl, pH 7.0
and reactions were started by the addition of NADPH (0.5 mM) with
other chemicals at concentrations specified in the Figure

Legends. The oxidation of NADPH was measured at 3M0 nm under
conditions identical to those utilized for iron release assays except
that NADPH was included at 0.2 mM.

Analysis of the Structure and Iron thake Ability of Ferritin
Followinngron Release by Paraquat

To assess the effect of iron release by paraquat on the protein
and its subsequent ability to reincorporate iron, reaction mixtures
containing ferritin (500 ug). NADPchytochrome P450 reductase (0.2 U),
paraquat (0.5 mM), ADP (5 mM), and NADPH (0.5 mM) in 50 mM NaCl, pH
7.0, (final volume 200 pl) were incubated either aerobically or
anaerobically for 30 min. In addition, one reaction mixture contained
no enzyme or paraquat, another contained enzyme but no paraquat, and
in one the iron was reduced using dithionite (1 mM) in place of enzyme
and paraquat. Following incubation the reaction mixtures were
chromatographed on Sephadex 6'25 (1 x 25 cm) in 0.1 M Hepes, pH 9.6.
One ml fractions were collected and aliquots (50 ul) were analyzed for
ferritin by ELISA as per Thomas and Aust (36). Appropriate fractions
were pooled and total protein determined by the BCA method (37).

Iron uptake by ferritin was assayed using the procedure of
Collawn gt El- (38) with minor modification. Incubation mixtures (5
m1 final volume, 0.1 M Hepes) contained aliquots of the pooled
ferritin fractions such that the final concentration of ferritin was

80 ug/ml and the pH of the incubations was adjusted to 6.8. The

128

reaction was started by the addition of ferrous ammonium sulfate (0.5
mM) and at 0, 2, A, and 6 min, 1 m1 aliquots of the reaction mixtures
were mixed with 0.3 g of Chelex 100 in a 13 x 100 mm test tube.
Following vigorous mixing to facilitate chelation of free ferrous
iron, the aliquots were transferred to a 1 ml plastic Eppendorf tube,
into which pin holes had been made, and which rested atop a 13 x 100
mm test tube. Low speed centrifugation was used to separate the
ferritin from the Chelex 100 and the absorbance of the recovered
protein at 310 nm was recorded as an index of incorporation of iron
into ferritin.

Evaluation of the nature of the ferritin protein structure
following iron release by paraquat, both aerobic and anaerobic, was
done using gel electrophoresis. Aliquots of reaction mixtures similar
to those utilized for the iron uptake studies were subjected to native
gel electrophoresis (5.“51 polyacrylamide) and SDSvgel electrophoresis
on a 15% polyacrylamide gel. The concentration of ferritin applied to
each lane was 7.5 ug and the gels were stained for protein using
Coomassie Blue. Ferritin not treated with paraquat was also run for
comparative purposes.

Paraquat Radical Detection

Formation of the paraquat cation radical resulting from NADPHv
cytochrome P050 reductase catalyzed reduction was monitored at 603 nm
(39) under anaerobic conditions with reactants as specified in the
Figure Legend. Detection of the paraquat radical in the presence and
absence of ferritin was also done under anaerobic conditions using ESR
spectroscopy (A0). Incubations were constituted at room temperature

and directly transferred to the cuvette of a Varian Century~112 ESR

129

spectrometer. Spectrometer settings were: 3320.0 0 magnetic field, 15
mw microwave power, 9.u275 GHz frequency, 1000 KHz modulation
frequency, 0.63 modulation amplitude, 0.5 sec time constant, A x 10"
gain, and a 100 G scan over A min. Concentrations of reactants are

specified in the Figure Legends.

RESULTS

Effect of SOD and 02 on Iron Release from Ferritin

The results in Table 1 demonstrate that the release of iron from
ferritin by NADPH-cytochrome PMSO reductase and paraquat is only
inhibited 62% by SOD. This is in contrast to the 971 inhibition,
observed when SOD is added to a system containing xanthine and
xanthine oxidase. A low rate of iron release was observed with
NADPchytochrome Pu50 reductase in the absence of paraquat and this
activity was almost completely inhibited by the addition of SOD.

Table 1. Effect of SOD on Xanthine Oxidase and Paraquatv
Dependent Release of Iron from Ferritin.

 

(nmol FeZIImin)
System vSOD + $00

Xanthine oxidase 0.6A 0.02

NADPchytochrome PH50
reductase 0.19 0.03

NADPH'cytochrome PHSO
reductase plus paraquat 0.66 0.25

 

Reaction mixtures (1ml final volume) contained rat liver ferritin
(1600 nmol Fe3+), bathophenanthroline sulfonate (1 mM), catalase (500
U) and, where indicated, xanthine (0.33 mM), xanthine oxidase (0.025
U), NADPH (0.5 mM), NADPchytochrome’PHSO'reductase (0.10 U), paraquat
(1 mM), and SOD (500 U) in 50 mM NaCl, pH 7.0. Reactions were
initiated by the addition of apprOpriate enzyme and continuously
monitored at 530 nm.

The data in Table 2 demonstrate that the rate of iron release
from ferritin isolated from the livers of control animals (native) is
much greater than that from ferritin obtained from animals pretreated

130

131

Table 2. Effect of SOD on Aerobic, ParaquatvDependent
Release of Iron from Ferritin.

 

(nmol Fe2+/min)
Ferritin (nmol Fe3+) vson +300

 

I. Native Ferritin

0.1 mg (9“) 0.21 0.0a
0.25 mg (235) 0.A2 0.07
0.50 mg (“70) 0.80 0.16
1.00 mg (9M0) 1.5“ 0.38
1.50 mg (1H10) 1.76 0.5”

II. IronvInduced Ferritin

0.1 m3 (320) 0.12 0.05
0.25 mg (800) 0.38 0.1a
0.50 mg (1600) 0.7u 0.29
1.00 mg (3200) 0.91 0.u5
1.50 mg (“800) 1.A“ 0.72

 

Reaction mixtures (1 m1 final volume) contained NADPchytochrome PHSO
reductase (0.10 U).‘NADPH (0.5 mM). paraquat (1 mM), catalase (500 U),
bathophenanthroline sulfonate (1 mM), varying amounts of rat liver
ferritin, and, where indicated, SOD (500 U) in 50 mM NaCl, pH 7.0.
Reactions were initiated with NADPH and continuously monitored at 530
nm.

132

with iron'dextran (iron-induced). However it can also be seen that,
even when the ferritin concentration was varied greatly,
paraquatvdependent iron release from both ferritin preparations was
consistently inhibited only 50'601 by SOD.

In Figure 1 it can be seen that when NADPH-cytochrome PHSO
reductase and paraquat were incubated with purified rat liver
ferritin, a lag period of approximately 7 min was observed before a
rapid release of iron from ferritin occurred, as indicated by
formation of the Fe2+vbathophenanthroline complex absorbing at 530 nm
(curve A). When the reaction mixture was purged with argon,to
decrease the 02 content prior to the addition of NADPH, the lag perbod
was shortened to approximately 3 min (curve B). When glucose oxidase,
glucose, and catalase were included to further reduce the
concentration of 02, iron release was observed immediately (curve C).
Effect of Varying Concentrations of Ferritin, NADPchytochrome PHSO
Reductase, and Paraquat on Iron Release from Ferritin by Paraquat.

The quantity of iron released by paraquat anaerobically was
assessed by continuously monitoring iron release from ferritin until
no further formation of the bathophenanthroline sulfonate - ferrous
complex was observed. The data in Table 3 demonstrate that all of the
iron was released from ferritin over a range of ferritin
concentrations when NADPchytochrome Pu50 reductase and paraquat were

incubated with ferritin anaerobically.

133

.A: opv mmmufixo mmoosaw cam .Azs my mucosam .AD Ptov mnmuosumg

omam meoncoou>o1mmo<z .Azs mm.ov pmsamnmq on cam Newman L4 can A: F.ov announce;

omza msogcooszo1xaa<z .Azs mm.ov umsamnma Amy “A: m.ov mmmuosumg omen osognoopao1mao<z
.Azs Fv umscmgmq A<v .s: omm um ecgouacos new Axe m.ov mmo<z Mo :oHuHunm 0:» an commanded
one: new 0.» ma .Homz :8 om ea As coo—V ommamumo cam Axe Pv mpmcomasm ocaaognucmcmnaonpmn
.A+mmm z: mmv capancmu Lm>«H um; cocamucoo Aoasao> Hanan as Pv.monsuxas

cofiuommm .umscmgmm an :Huungmm song :OLH uo mmmoamm co cmmhxo ho vacuum .F ogsmwm

134

 

 

 

 

 

 

 

10

 

 

TIME (min)

135

Table 3. Anaerobic Iron Release from Ferritin by Paraquat at
Varying' Ferritin Concentrations.

 

 

Ferritin Fe3+ Added (nmol) Total Fe3+ Reduced (nmol)
O 0.21
25 26.1
50 50.“
75 73.9

100 96.1

 

Reaction mixtures (1 ml final volume) contained NADPchytochrome PASO
reductase (0.1 U). paraquat (0.25 mM), bathophenanthroline sulfonate
(1 mM). glucose (5 mM). glucose oxidase (10 U), catalase (500 U), and
varying amounts of rat liver ferritin as indicated in 50 mM NaCl, pH
7.0. Reactions were initiated by the addition of NADPH (0.5 mM) and
continously monitored at 530 nm until no further formation'of the
bathophenanthroline sulfonate-Fe2+ complex was observed. Results are
the averages of two separate experiments. '
As shown in Figure 2, the time required to release all of the iron
from ferritin was decreased by increasing NADPchytochrome P950
reductase activity. A marked decrease in the amount of time required
for complete release of iron was also observed when the concentration
of paraquat was increased, up to approximately 0.10 mM paraquat
(Figure 3).
Release of Ferritin Iron and NADPH Oxidation by Other Bipyridyls
Other bipyridyls which undergo one electron reduction were also
investigated for their ability to mediate the release of iron from

ferritin under anaerobic conditions. As shown in Table A diquat and

benzyl viologen, which are also reduced to cation radicals, also

136

Figure 2. Effect of Varying NADPH-Cytochrome P950 Reductase
Activity on the Time Required for Complete Release of Iron from
Ferritin by Paraquat. Reaction mixtures (1 ml final volume)
contained rat liver ferritin (25 uM Fe3+),‘paraquat (0.25 mM),
bath0phenanthroline sulfonate (1 mM), catalase (1000 U), glucose
(5 mM), glucose oxidase (10 U) and'varying amounts of
NADPchytochrome P950 reductase as indicated in 50 mM NaCl, pH
7. 0. Reactions were initiated by the addition of NADPH (0. 5 mM)
and continously monitored to completion at 530 nm.

137

 

 

 

_ _
5 m

AEEV ZOFMJQZCU O... 92:.

 

.I5 .20 .25
REDUCTASE (U/ml)

.IO

.05

138

Figure 3. Effect of Varying Paraquat Concentration on the Time
Required for Complete Release of Iron from Ferritin by Paraquat.
Reaction mixtures (1 ml final volume) contained rat liver
ferritin (25 uM Fe3*), NADPchytochrome PN50 reductase (0.1 U),
bathophenanthroline sulfonate (1 mM), catalase (1000 U), glucose
(0.5 mM), glucose oxidase (10 U) and varying amounts of paraquat
as indicated in 50 mM NaCl, pH 7.0. Reactions were initiated by
the addition of NADPH (0.5 mM) and‘continously monitored to
completion at 530 nm. ‘

139

 

 

 

 

- b

 

 

20"

w m
Eevzofimdzoo 8. ms;

.20

.15

.IO
PARAQUAT(mM)

.05

140

released iron from ferritin at a rapid rate. The data presented were
calculated from the initial linear portion of the curve, such as is
shown in Figure 1 (curve C).

Table A. Release of Iron from Ferritin and NADPH Oxidation by
Redox Active Bipyridyls.

 

 

 

iron released NADPH oxidation nmol NADPH oxidized

compound (nmol/min) (nmol/min) nmol iron released
Paraquat A.88 11.58 2.37
Diquat 5.05 12.22 2.A2
Benzyl Viologen 3.A3 9.00 2.62

 

Reaction mixtures for iron release (1 m1 final volume) contained
NADPchytochrome PASO reductase (0.1 U), rat liver ferritin (25 uM
Fe 3+), bathophenanthroline sulfonate (1 mM), glucose (5 mM), glucose
oxidase (10 U), catalase (500 U), and the various bipyridyls (all at
0.25 mM) in 50 mM NaCl, pH 7.0. Reactions were initiated by the
addition of NADPH (0. 5 mM) and monitored continously at 530 nm. For
NADPH oxidation studies the chelator was omitted, NADPH was 0. 2 mM,
and the decrease in absorbance at 3A0 nm was monitored.
The rates of NADPH oxidation in the presence of the various compounds
were also determined and compared to the rates at which the chemicals
catalyzed the release of iron from ferritin. These data indicated
that approximately two nmol of NADPH were oxidized per nmol of iron
released for the three chemicals tested.
Effect of Ferritin on Detection of the Paraquat Radical

As shown in Figure A, as little as 8 ug/ml of ferritin (25 uM
Fe3*) reduced the 603 nm absorbance of the paraquat radical (curve B).

When the amount of ferritin was continually increased, up to 6A ug/ml

protein containing 200 uM Fe3+, (curves CvE), a lag period proceeding

141

 

.aauaggwu 26>HH um; A+mmm z: oomv m: as

Amy use .A+mmm :3 00.2 m: mm ADV .A+mmm :1 omv m: o_ “as .A+mmm :: mmv m: w Amy .mcofluaeem
on A<v .sc moo pm cmnouucos mamsoscaucoo ucm Axe my mmo<z cofiufinum on» ma uopmwuHca

one: mogapxaa cofiuomom .Homz zs om CH umpmofiucw mm :HuHLLou Lm>HH am; no mucsosm mcamgm>
can A: opv emanaxo omoosaw .Aze my omoosaw .A: oomv onmamumo .Aza mm.ov umzummma «A: —.ov
ommposumn omnm msognoou>otmmo<z uwcfimpcoo Amesao> Hanan as Pv mogsuxas coHuomom ..Hmoaumm
cofipmo umzcmumm on» no cofipoopoa ofinumsoponaoguomam :o caufingmm uo vacuum .z onsmam

142

 

 

 

 

 

 

 

 

(WUEO9MJJSN30 ‘1VOIldO

(D
9
’3
9.?
DJ
2
g...
“IO
«0
1 1 I l l
o (I) to <1- N
—. O O O O

144

.A+mmm z: oomv m: =0 Amv new uA+mom :3 oopv m: mm “av ua+mom :3 omv mg mp

on nfl+mmm z: mmv m: m Amv uncofiuaunm o: A<v uncofiumnpcmocoo mcfizoaaou on» um concave“ mm:
cfiuagnmu Lm>HH pmm, .mnonumz new mamagmpmz CH cocdauso mm ncoapfivcoo mmm no“: hHoumHumesfi
umccmom cam A28 m.ov mmo<z no acquficcm on» >9 coumupfi:«.ouoz neoHuommm .0.» ma .Homz

:8 om.cH A: o_v mmmofixo whoosam cam .Aza mv omoosam .Azs mm.ov uncommon .A: oomv ommampmo
.AD F.ov commence; omzm osonnoou>o1mma<z oocfimucoo Amasao> Hagan Ha PV momspxas_coauomom
.mmm an amusemm cofiumo panamama an» no coauompma co caufiagmm do vacuum .m mesmam

146

\l

.nvosumz cam namwnoum: a“ nocaauso mm acoHuHucoo mmm so“: mam>flufiuoamp eczema» can Axe
m.ov mmo<z uo coauaucm on» up cmumquaca one: ncOauommm .A+mom z: omv cauuguou Lm>«H.umL
new A: opv mnmuwxo omoosaw .Azs my onoosam .Aza mm.ov unscmpmn .Aa oomv onmamumo .Aa _.ov
composed; omzm msonnoOpho1mma<z omcamucoo Aoesao> Hmcuu Ha Pv megapxaa coHuomom. .mmm an
:oHuoouma Hmowumm coHumo pascmnmm co :HuHLme no uoouum on» go cmom o>fluauonom .o mcsmwm

147

'1'

mo.

.JL

0.

25
m2;

148

Figure 7. Iron Uptake by Ferritin. Reaction mixtures (5 ml
final volume) contained ferritin (80 ug/ml) in 0.1 Hepes, pH 6. 8.
Reactions were started by the addition of ferrous ammonium
sulfate (0. 5 mM) and quenched in 0. 3 g Chelex 100 as described in
Materials and Methods with the absorbance at 310 nm recorded as
an index of iron uptake. The ferritin preparations were
recovered from incubation mixtures detailed in Materials and
Methods and are as follows: 0 = no ferritin, Fe2+ only; 0 = no
treatment; X = NADPchytochrome PA50 reductase; Ia
NADPchytochrome PA50 reductase and paraquat, aerobically; 4A=
NADPchytochrome PA50 reductase and paraquat, anaerobically: 1A=
dithionite.

 

 

149

 

 

0.4-

a .2
O O
“E: 9» <43 mx<kdn 20m.

0.l-

 

 

TIME (min)

150

non'treated ferritin sample. Likewise, the individual subunits
exhibited mobility identical to that of the nonvtreated ferritin and
no subunit fragments of less than approximately 18,000 molecular

weight were detected (results not shown).

DISCUSSION

The results of this study demonstrate that free radicals produced
by NADPchytochrome PA50 reductase catalyzed reduction are capable of
releasing iron from ferritin. The toxicities associated with these
compounds is often attributed to their ability to undergo redox
cycling resulting in the production of active oxygen species leading
ultimately to oxidative damage of cellular constituents, including
membrane unsaturated lipids, proteins and DNA (2). In general, it
appears that oxidative damage requires, or is greatly potentiated by,
the presence of transition metals therefore the identification of
biological sources of iron is critical to an understanding of the
mechanism(s) governing the toxicity of these chemicals. We had
previously demonstrated the ability of 02:, generated by xanthine
oxidase, to release iron from ferritin (18) suggesting that increased
production of 02? in biological systems may contribute to oxidative
stress by increasing the amount of iron available to participate in
redox reactions. Similarly, the redox cycling of paraquat promoted
ferritinvdependent lipid peroxidation (20).

Xanthine oxidase-dependent iron release was inhibited 971 by SOD
showing that the enzyme can prevent essentially all of the
027-dependent iron release. However, the present data demonstrate
that under aerobic conditions SOD only partially prevents
paraquatvdependent release of iron from ferritin. Iron release from
ferritin isolated from animals not pretreated with iron was even
greater, as has been previously demonstrated using reduced flavins

(A1). This suggests that iron release from ferritin may be even more
‘ ’ 151

152

significant in 1119. The difference between total iron release and
SODvsensitive iron release was used to assess the Ozfvindependent
release of iron and indicated that AOvSOS of the iron released could
be attributed to reduction by a reductant other than 027.

The direct reduction of Fe3+ by the paraquat radical has been
previously reported. (27,28) Recent work has indicated that the
reduction potential required for ferritin iron release is
approximately -230 mV at pH 7.0 (17), therefore the paraquat radical
(vAAA mV) (39) may be more effective at mediating the reductive
release of iron from ferritin than 02? (-330 mV) (A2). The second
order rate constant for the reaction of the paraquat radical with
oxygen is 8 x 108 M’1 sec’? (A3) and 2.1 x 108 M'1 sec’? with
cytochrome c (28). In these studies, the ferric iron within ferritin
appears to compete effectively with oxygen for the radical as
evidenced by the ability of the radical to release nearly 501 of the
iron from ferritin even under aerobic conditions.

In agreement with the above data, the results in Figure 1
demonstrate the efficiency with which the paraquat radical can release
ferritin iron. Under aerobic conditions a very low rate of iron
release occurs. However, at the onset of anaerobiosis, resulting from
continued redox cycling (at approximately 7 min), a rapid rate of iron
release was observed. Accordingly, lowering the oxygen concentration
of the reaction mixture considerably shortened the time required to
achieve rapid iron release while anaerobiosis at the initiation of the
reaction resulted in an immediate release of iron from ferritin.
Also, reductive release of ferritin iron by the paraquat radical was

much more complete than with 027, as virtually 1001 of the iron could

153

be removed, whereas 02? appeared to release only approximately 5} of
the iron from purified rat liver ferritin (18). The dependency of the
time required for complete release of iron from ferritin on both the
activity of NADPchytochrome PA50 reductase and the concentration of
paraquat provide further evidence that iron release was mediated by
the paraquat radical.

The remarkable ability of the paraquat radical to effectively
release iron from ferritin suggested that perhaps other bipyridyl
radicals would be similarly capable of mobilizing iron from ferritin.
Accordingly, our data demonstrate that the cation radicals of benzyl
viologen and diquat were also very effective at mediating release of
iron from ferritin. It is difficult to assess the effectiveness of
the various radicals to release ferritin iron solely on the basis of
rates of iron release due to varying rates of reduction of the
chemicals by the enzyme, therefore rates of NADPH oxidation were also
monitored. NADPH oxidation rates in the presence of the enzyme and
the chemicals are an indicator of the rate of formation of the radical
form of the chemicals. The results demonstrated that two nmol of
NADPH were oxidized per nmol iron released, thus the process was not
tightly coupled. This finding is not totally surprising as iron
release from ferritin is highly dependent upon a number of factors
including redox potential and size of the reductant (15), complex
formation at the iron core surface (16), and the amount of surface
area on the iron core available for release (13). Conversely,
paraquat can undergo div or tetravalent reduction or even
disproportionation (39) which would significantly affect NADPH

oxidation.

154

The rapidity at which the paraquat cation radical mediates the
complete release of iron from ferritin suggests that electron transfer
from the radical to the ferric hydroxide iron core, or to
oxidovreduction sites on the protein, occurs very rapidly under
anaerobic conditions. Confirmation of this was provided by the ESR
and UVvvisible spectroscopy studies used to detect paraquat radical
formation. The addition of as little as 8 ug/ml of ferritin protein
(25 pH Fe3*) resulted in a marked diminution of the ESR free radical
signal of paraquat, with no detectable signal at 6A ug/ml of ferritin.
Similar data were obtained when paraquat radical formation was
monitored at 603 nm. As the incubations were monitored over time the
paraquat radical could be detected once the iron was released from
ferritin. Accordingly, if the ESR reaction mixtures
were scanned repetitively over time, a reappearance of the signal was
observed as the iron was released from ferritin, thereby allowing an
accumulation of the paraquat radical. These data indicated that there
was significant interaction between the radical form of paraquat and
the protein and that loss of the signal was likely due to transfer of
electrons from the cation free radical to ferritin.

To provide further evidence that paraquatvdependent release of
iron from ferritin occurs by simple electron transfer, and does not
alter the protein, gel filtration chromatography was used to recover
ferritin following its incubation with paraquat and NADPchytochrome
PA50 reductase, under aerobic and anaerobic conditions. Enzyme linked
immunosorbent analysis of fractions collected from the Sephadex 0'25
column revealed essentially no change in the reactivity of rabbit,

antivrat liver ferritin IgG towards the ferritin (results not shown).

155

In agreement, native and denaturing gel electrophoresis of the
ferritin demonstrated there was no effect on the assembled protein or
on the individual subunits (data not shown).

It is known that ferritin does have sites sensitive to
degradation by proteolysis (AA) and pH (A5). Preliminary experiments
in this laboratory have indicated that ironvfree ferritin, prepared by
incubation of ferritin with the paraquat radical, was much more
susceptible to proteolysis. These results agree with other studies
(AA,A6) which reported that apoferritin is more readily degraded.
Therefore, while the paraquat radical may not directly degrade the

protein, paraquat intoxication in vivo may lead to disruption of the

 

normal ironvstorage capacity of cells.

Additionally, iron uptake by ferritin was determined to confirm
that the biologic function of ferritin was not compromised by radical
mediated iron release. Iron uptake was greatest for ferritin from
which the most iron was released (the paraquat radical under anaerobic
conditions or by dithionite). These results confirm previous work
(13) which demonstrated that iron uptake was most rapid for those
molecules containing the least iron. The data presented in Figure 6
have been corrected to account for the iron present in ferritin prior
to addition of ferrous iron, thus while iron uptake was greatest for
ferritin containing the least iron, by 6 min all ferritin preparations
contained nearly the same amount of iron. Subsequently, the iron in
all of the reloaded ferritin preparations could be reductively
released by 027 or the paraquat radical (data not shown).

These results provide evidence that the generation of organic

free radicals in biological systems can serve to mobilize cellular

156

iron from its storage protein. This finding may help to explain the
variable results obtained when using SOD to protect against paraquat
intoxication (25,26) i.e., that the relative contributions of 02? and
the paraquat radical to paraquat toxicity may be highly dependent upon
oxygen concentrations within the cell as has been proposed for carbon
tetrachloride (A7,A8). It is conceivable that continuous redox
cycling of the herbicide may lead to hypoxic conditions which could
favor the rapid release of ferritin iron by the paraquat radical while
still maintaining a concentration of oxygen sufficient to promote
lipid peroxidation. In this regard, it has been reported that hypoxia.
may potentiate paraquat toxicity 12 1112 (2A). Also, Cadenas 21 31.
(A9) observed a burst of chemiluminescence upon oxygenation of
microsomes previously incubated anaerobically with NADPH and paraquat.
The intensity of chemiluminescence was related to the amount of
paraquat radical accumulated during anaerobic incubation.

Barry (50) has reported that human liver iron stores (ferritin)
range from 10.8v1A.3 mM Fe3*, thus it is conceivable that interaction
between the radical and the protein may occur 12 1112. However, the
lung is the primary target organ of paraquat toxicity, presumably due
to an active uptake mechanism (51). Ferritin is present in almost all
tissues at varying levels (52) but the present studies indicate that
as little as 8 us (25 0M Fe3*) of ferritin can readily accept
electrons from the paraquat radical, therefore similar interactions in
the lung are feasible. Another factor to be considered with respect
to physiologic significance is the subcellular distribution of
ferritin. While ferritin is found predominantly in the cytosol,

electron microscopy studies have demonstrated the existence of

157

ferritin in the endoplasmic reticulum (53) and, accordingly, ferritin
is found in liver (36) and lung (C.E. Thomas and S.D. Aust,
unpublished) microsomes prepared by differential centrifugation at
105,000 x g for 90 minutes. Very recent work indicates that ferritin
bound iron remains tightly associated with the microsomal fraction
even following sucrose gradient centrifugation (5A). Therefore, it is
likely that, in 1119, ferritin is in close proximity to the site of
paraquat radical and 027 generation on the endoplasmic reticulum and
that little diffusion of these radical species through the cellular
millieu to access ferritin is required.

While the biochemical mechanisms governing the toxicities of
paraquat and other redox active chemicals are likely to be very
diverse, these data suggest that release of iron from ferritin may be
a common feature contributing to free radical mediated toxicities. It
has been recently demonstrated that iron greatly potentiates the
toxicity of paraquat in mice (55). If this iron is derived ultimately
from ferritin remains to be determined, however. These findings would
indicate that the administration of appropriate iron chelators may
help to ameliorate the toxicity of these compounds. In agreement,
EDTA and desferrioxamine prevented paraquat induced bacterial
inactivation (56) and desferrioxamine treatment prolonged survival of
paraquat treated mice (55). These results suggest that redox active
metals such as iron may play an integral role in paraquat toxicity.
Thus, the demonstration that the free radicals of paraquat and related
toxic herbicides possess the capability to delocalize tissue iron may
provide insight into the mechanisms which are involved in the
toxicities of these compounds and lead ultimately to more effective

prevention and treatment regimens.

10.

11.

12.

13.

LIST OF REFERENCES

H. Kappus, and H. Sies. Toxic drug effects associated with
oxygen metabolism: redox cycling and lipid peroxidation.
Experentia 37: 1233*12A1 (1981).

B. Halliwell and J.M.C. Gutteridge. Lipid Peroxidation: a
radical chain reaction. In: Free Radicals in Biology and

 

Medicine, pp. 139~189,'Claredon Press, Oxford (1985).

D.T. Sawyer and J.S. Valentine. How super is superoxide?
Accts. Chem. Res. 11: 393~A00 (1981).

S. D. Aust, L. A. Morehouse, and C.E. Thomas. Role of metals in
oxygen radical reactions. J. Free Rad. Biol. Med. 1: 3'25
(1985). - ‘ - ~ - -

A. Jacobs. Low molecular weight intracellular iron transport
compounds. Blood 59: A33vA39 (1977).

G.R. Bartlett. Phosphate compounds in rat erythrocytes and
reticulocytes, Biochem. BiOphys. Res. Commun. 19: 1055*1062
(1976). ‘ ' ' ' ‘ ‘

S. Pollack, T. Campana and J. Weaver. Low molecular weight
iron in guinea pig reticulocytes. Amer. J. Hematol. 12: 75'8A
(1985). ' ' ' ‘

K.L. Fong, P.B. McCay, P.B., J.L. Poyer, H.P. Misra, and 8.8.
Keele. Evidence for superoxidevdependent reduction of Fe3
and its role in enzyme-generated hydroxyl radical formation.
Chem. Biol. Inter. 15: 77-89 (1976)

C. G. D. Morley and A. Bezkorovainy. Identification of the iron
chelate in hepatocyte cytosol. IRCS Med. Sci. Libr. Compend.
11: 1106-110? (1983). . . . . .

P. Aisen and I. Listowsky. Iron transport and storage

proteins. In: Annual Review of Biochemistry, Volume A9 (E.E.
Snell, P. D. Boyer, A. Meister and C. C. Richardson, eds.) pp.
357'393 (1980). ‘ '

 

P.M. Harrison. Ferritin: An iron'storage molecule. Sem.
Hematol. 11: 55'70 (1977). ‘ '

E.C. Theil. Ferritin: Structure, function and regulation.
In:' Iron Binding Proteins Without Cofactors or Sulfur

 

Clustersg(E. C. Theil, G. L. Eichhorn, and L. G. Marzilli, eds. .)’

pp. 1'38, Elsevier, New York (1983).

P.M. Harrison, T.C. Hoy, I.G. Macara, and R.J. Hoare.
Ferritin Iron Uptake and Release. Structurevfunction
relationships. Biochem. J. 1A3:'AA5'A51 (197A).

158

1H.

15.

16.

17.

18.

19.

20.

21.

22.

23.

2H.

25.

26.

159

C.P.A. Bryce and R.R. Crichton. The catalytic activity of
horse spleen apoferritin. Preliminary kinetic studies and the
effect of chemical modification. Biochem. J. 133: 301'309
(1973) ' ' ' ‘

T. Jones, R. Spencer, and C. Walsh. Mechanism and kinetics of
iron release from ferritin by dihydroflavins and dihydroflavin
analogues. Biochemistry 11: N011vu017 (1978).

F. Funk, J. P. Lenders, R. R. Crichton, and W. Schneider.
Reductive mobilization of ferritin iron. Eur. J. Biochem.
152: 167 172 (1985). . - .

G. D. Watt, R. B. Frankel and C.C. Papaefthymiou. Reduction of
mammalian ferritin. Proceed. Natl. Acad. Sci. gg: 36u0'36h3
(1985) ' ' ‘ ' ‘

C. E. Thomas, L. A. Morehouse, and S. D. Aust. Ferritin and
superoxidevdependent lipid peroxidation. J. Biol. Chem. 260:
3275:3280 (1985). ' ' '

P. Biemond, H.G. Van EiJk, A.J.G. Swaak, and J.F. Koster.
Iron mobilization from ferritin by superoxide derived from
stimulated polymorphonuclear leukocytes. J. Clin. Invest.
73: 1576' 1579 (198”). ' ' ‘

M. Saito, C. E. Thomas, and S. D. Aust. Paraquat and
ferritinvdependent lipid peroxidation. J. Free Rad. Biol. Med.

1: 179'185 (1985)-

J.S. Bus, S.D. Aust, and J.B. Gibson. Superoxidev and singlet
oxygenvcatalyzed lipid peroxidation as a possible mechanism
for paraquat (methyl viologen) toxicity. Biochem. Biophys.
Res. Comm. 58: 7N9*755 (197“). ‘ '

H.M. Hassan, and I. Fridovich. Superoxide radical and the
oxygen enhancement of the toxicity of paraquat in Escherichia

 

Coli. J. Biol. Chem. £23: 81u3~81fl8 (1978).

R.S. Simona, P.S. Jackett, M.B.W. Carrol, and D.B. Lowrie.
Superoxide independence of paraquat toxicity in Escherichia

 

coli. Toxicol. Appl. Pharmacol. 31: 271'280 (1976).

L.L. Smith, and M.S. Rose. Biochemical changes in lungs
eXposed to paraquat. In: Biochemical Mechanisms of Paraquat

Toxicity (A. P. Autor, ed. ), pp. 187v200, Academic Press, New

York (1977).

A.P. Autor. Reduction of paraquat toxicity by superoxide
dismutase.‘ Life Sci. 13: 1309*1319 (197”).

C. E. Patterson, and M. L. Rhodes. The effect of superoxide
dismutase on paraquat mortality in mice and rats. Toxicol.
Appl. Pharmacol. 62: 65'72 (1982). '

27.

28.

29.

30.

31.

32.

33.

3".

35.

36.

37.

160

H.G. Sutton and C.C. Winterbourn. Chelated iron-catalyzed fOH
formation from paraquat radicals and H202: Mechanism of
formate oxidation. Arch. Biochem. BiOphys. 235: 106-115
(198“). ' ‘ ' ' “

E. J. Land and A. J. Swallow. Electron transfer from pyridinyl
radicals to cytochromevc. 'C. Ber. Bunsenges. Phys. Chem. 12;
u36-u37 (1975). ' ' ‘ ’ ‘

R. F. Beers and I. w. Sizer. A spectrophotometric method for
measuring the breakdown of hydrogen peroxide by catalase. J.
Biol. Chem. 195: 133-1uo (1952). ‘

J. M. McCord and I Fridovich. Superoxide dismutase. An
enzymic function for erythrocuprein (hemocuprein). ‘ J. Biol.
Chem. 2AM: 60u9v6055 (1969) '

T.C. Pederson, J.A. Buege, and S.D. Aust. Microsomal electron
transport. The role of reduced nicotinamide adenine
dinucleotide phosphatevcytochrome c reductase in liver
microsomal lipid peroxidation. J. Biol. Chem. gig: 713uv71u1

Y. Yasukochi and 8.3.8. Masters. Some properties of a
detergentvsolubilized NADPchytochrome c (cytochrome p-u50)
reductase purified by biospecific affinity chromatography. J.
Biol. Chem. 251: 5337'53hfl. (1976). ‘

J.W. Halliday. Immunoassay of ferritin in plasma. In:
Methods in EnzymolognyVolume 8“ (L Langone and L

 

VanVunakis, eds.) PP. 1H8v171, Academic Press, New York
(1982)

P.B. Brumby and V. Massey. Determination of nonheme iron,
total iron, and cOpper. In: Methods In EngymologyL4Volume 10
(R. W. Estabrook and M. E. Pullman, eds.) pp. N63vu7h, Academic
Press, New York (1967).

P. Carter. Spectophometric determination of serum iron at the
submicrogram level with a new reagent (Ferrozine). Anal.
Biochem. LO: HSOvH58 (1971L '

C.E. Thomas and S.D. Aust. Rat liver microsomal
NADPHvdependent release of iron from ferritin and lipid
peroxidation. J. Free Rad. Biol. Med. 1: 293*300 (1985).

M.C. Redinbaugh and R.B. Turley. Adaptation of the
bicinchoninic acid protein assay for use with microtiter
plates and sucrose gradient fractions. Anal. Biochem. l§§3
267'271 (1986). ' ' '

38.

39.

HO.

1‘1.

”2.

“3.

nu.

HS.

”6.

H7.

”8.

H9.

50.

161

J.F. Collawn, Jr., D.G. Priest and W.H. Fish. A sensitive

assay employing a solidvphase metal"chelator'to estimate rates
of iron accumulation by ferritin. Anal. Biochem. 127: 105'111
(1982). . . . . . ...

A. Ledwith. Electron transfer reactions of paraquat. In:
Biochemical Mechanisms of Paraquat Toxicity (A. P. Autor, ed. ),
pp. 21'38, Academic Press, New York (1977) '

R. P. Mason and J. L. Holtzman. The role of catalytic
superoxide formation in the 02 inhibition of nitroreductase.
Biochem. Biophys. Res. Commun. 67: 1267 127A (1975).

J.R. Mertz and E. C. Theil. Subunit dimers in sheep spleen
apoferritin. The effect On iron storage. J. Biol. Chem. 258:
11719-11726 (1983) ' ' "

J.A. Fee and J.S. Valentine. Chemical and physical properties
of superoxide.' In: Superoxide and Superoxide Dismutases

(A. M. Michelson, L M. McCord and I. Fridovich, eds. ), pp.
19‘60, Academic Press, New York (1977).

P. S. Ras and E. Hayon. Redox potentials of free radicals.
IV. Superoxide and hydrOperoxy radicals -02’ and -H02. J.
Phys. Chem. 79: 397vu02 (1975). ‘

R.R. Crichton. Studies on the structure of ferritin and
apoferritin from horse spleen II. Chymotrypsin, subtilisin,
cathepsin D and pepsin digestion of ferritin and apoferritin.
Biochim. Biophys. Acta 229: 75'82 (1971)

R.R. Crichton and C. F. A. Bryce. Subunit interactions in horse
spleen apoferritin. Dissociation by extremes of pH. Biochem.
J. 133: 289-299 (1973) "

G. C. Wood and R. R. Crichton. Optical rotary dispersion and
circular dichroism studies on ferritin and apoferritin.
Biochim. Biophys. Acta 229: 83-87 (1971).

H. Kieczka, R. Remmer, and H. Kappus. Influence of oxygen on
the inhibition of liver microsomal activation of carbon
tetrachloride by the catechol 2~hydroxy estradiolv178.
Biochem. Pharmacol. 30: 319v32h (1981L

H. Kieczka and H. Kappus. Oxygen dependence of CCluvinduced
lipid peroxidation in vitro and in vivo. Toxicol. Lett. g;
191-196 (1980). ' '

E. Cadenas, R. Brigelius and H. Sies. Paraquat'induced
chemiluminescence of microsomal fractions. Biochem.
Pharmacol. 32: 1M7~150 (1983). ‘

M. Barry. Progress Report. Iron and the liver. Cut. 12:
32u-33u (197M). ' '

51.

52.

53.

5”.

55.

56.

162

M. 8. Rose, L. L. Smith and J. Hyatt. Evidence for
energyvdependent accumulation of paraquat into rat lung.
Nature (London) 252: 31u—315 (197“)

R.R. Crichton. Ferritin. Struc. Bond. 11: 67*139 (1973).

G.R. Richter. The ironvloaded cell - The cytopathology of
iron storage. Amer. J. Pathol. 21: 363'396 (1978).

JvC. Sibille, JvN. Octave, YvJ. Schneider, A. Trovet, and R.R.
Crichton. Subcellular localization of transferrin protein and
iron in the perfused rat liver. Effect of Triton HR 1339,
digitonin and temperature. Eur. J. Biochem. 122: 7u'55

(1986). ' ' ~ -

R. Kohen and M. Chevion. Paraquat toxicity is enhanced by
iron and reduced by desferrioxamine in laboratory mice.
Biochem. Pharmacol.3H:18fl1-1893 (1985).

R. Kohen and M. Chevion. Transition metals potentiate
paraquat toxicity. Free Rad. Res. Commun. 1: 79-88 (1985).

CHAPTER IV

RELEASE OF IRON FROM FERRITIN BY CARDIOTOXIC
ANTHRACYCLINE ANTIBIOTICS

163

ABSTRACT

The use of the extremely effective anthracycline antitumor drugs,
adriamycin and daunomycin, is limited by a severe, dose-dependent carv
diomyOpathy. Anthracycline induced toxicity has been prOposed to inv
volve ironvdependent oxidative damage to biological macromolecules yet
little is known regarding the availability of physiologic iron. We
now report that, in the presence of NADPchytochrome PnSO reductase,
these drugs undergo redox cycling to generate 027 which mediates a
slow, reductive release of iron from ferritin, the maJor intracellular
iron storage protein. Anaerobically, the semiquinone free radical
forms of adriamycin and daunomycin catalyze a very rapid, extensive
release of iron from ferritin. In contrast, diaziquone, an aziridinyl
quinone antitumorigenic agent which is less cardiotoxic, is unable to
release iron from ferritin. Thus, the present studies suggest that
the cardiomyopathy observed with the anthracyclines, and perhaps their
antineOplastic activity as well, may be related to their ability to
delocalize tissue iron, thereby contributing to the formation of

strong oxidants capable of damaging critical cellular constituents.

164

INTRODUCTION

Adriamycin and daunomycin are members of a class of structurally
related antibiotics collectively known as anthracyclines. They are
widely used in chemotherapeutic treatment of cancer as their antineov
plastic properties encompass a wide range of tumors including those
associated with breast, lung, and soft tissue (1). Unfortunately, the
clinical use of these very effective antineoplastics is severely
limited by a dosevdependent cardiomyOpathy leading to congestive heart
failure (2). Total dosage of adriamycin must be limited to 550 mg/m2
body surface as patients receiving a total dose of greater than 600
mg/m2 had a H11 incidence of cardiomyopathy with approximately half of
these cases being fatal (3). The mechanism(s) governing the cytotoxic
and/or the antitumorigenic effects of the anthracyclines remain(s) to
be elucidated, however present evidence suggests that reduction of the
drugs to their corresponding semiquinone radicals by nuclear and micro~
somal flavoproteins (",5) is prerequisite for their cytotoxicity. The
resultant radical is then proposed to promote the oxidation of
cellular constituents such as membrane unsaturated lipids and/or DNA
via its ability to react with dioxygen to produce active oxygen
species such as 027 and H202 peroxide (6).

Recent evidence indicates that an interaction with iron is imporv
tant in causing oxidative damage to both DNA and unsaturated lipids by
these drugs (7). However, the significance of these findings is
limited by a lack of knowledge concerning the availability of iron in
1119. The majority of intracellular iron is stored within ferritin, a

large multivsubunit protein in which iron (up to 4500 atoms/molecule)

165

166

is stored as a ferric hydroxide core (8). Release of iron from
ferritin appears to require reduction of the ferric iron and, in
accordance, we have previously demonstrated the ability of 027,
generated by xanthine oxidase, to reductively release iron from ferriv
tin (9). The prOpensity of the anthracycline antibiotics to undergo
cyclic reduction and autoxidation to generate 02? led us to investiv

gate their ability to catalyze the release of iron from ferritin.

MATERIALS AND METHODS

Materials

Bathophenanthroline sulfonate, Mv7vdiphenyl'1,10'phenanthroline,
NADPH (Type III), cytochrome c (Type IV), Sephadex G~25, xanthine, and
.22222’ phenanthroline were purchased from Sigma Chemical Company (St.
Louis, MO). Glucose and ferric chloride were obtained from Baker
Chemical Company(Phillipsburg, NJ) and thioglycolate from Fisher
Chemical Company (Fairlawn, NJ). EDTA was a product of Mallinckrodt
Chemical Company (Paris, KY) while Ultragel AcA an was purchased from:
LKB (Bromma, Sweden), Chelex 100 from BiovRad Laboratories (Richmond,
CA) and Imferon from Merrell Dow (Cincinnati, OH). Diaziquone was a
generous gift of the Drug Synthesis and Chemistry Branch, Division of
Cancer Treatment, National Cancer Institute (Bethesda, MD) while
adriamycin was kindly supplied by Adria Laboratories (Columbus, OH).
Enzymes

Glucose oxidase (EC 1.1.3.M), xanthine oxidase (EC 1.2.3.2), and
bovine erythrocyte SOD (EC 1.15.1.1) were obtained from Sigma and
catalase (EC 1.11.1.6) was purchased from Millipore (Freehold, NJ).
Glucose oxidase contained contaminating iron (2.7 nmol Fe3*/Unit of
activity) and was therefore incubated with 10 mM EDTA for one hour on
ice and chromatographed over a Sephadex G~25 column equilibrated with
0.3 M NaCl, pH 7.0 to remove adventitious iron. Catalase was also
chromatographed on Sephadex G-25 and was assayed as per Beers and
Sizer (10) and xanthine oxidase was similarly treated to remove
ammonium sulfate and activity measured by aerobic reduction of

cytochrome c (11). The activity of SOD was also measured by the

167

168

method of McCord and Fridovich (11). Protease solubilized
NADPH-cytochrome Pu50 reductase (EC 1.6.2.h) was purified from
microsomes isolated from the livers of male SpraguevDawley rats
(250-300 g), previously pretreated with 0.1% phenobarbital in their
drinking water for 10 days, as previously described (12) with the
modification that enzyme eluting from the Sephadex G~100 column was
subjected to affinity chromatography on 2',5'vADP~Sepharose using
buffers containing no detergent. The final preparation of
NADPchytochrome PHSO reductase was chromatographed on a Sephadex G-25
column with 0.3 M NaCl, pH 7.0 Just prior to use, assayed, and
activity defined as 1 nmol cytochrome c reduced/min/ml.
Preparation of Ferritin and Iron Release Assays

Ferritin was purified from the livers of rats given an
intraperitoneal injection of Imferon (ironvdextran) as described by
Halliday (13) with modification (9). In addition, Ultragel AcA nu was
substituted for Sepharose 68. Following purification the ferritin was
incubated on ice in 10 mM EDTA for one hour and chromatographed on
Sephadex Gv25 to remove any adventitious iron. The total iron content
of ferritin was determined essentially as described by Brumby and
Massey (1A) with minor modification (9) and found to contain 3.12 pH
Fe/ug protein.

Iron release from ferritin was determined according to Thomas 32_
‘gl. (9) with modification. 'Qgthgvphenanthroline was used as the
ferrous iron chelator and the increase in absorbance at 510 nm was
continuously monitored in a Cary 219 dual beam recording
spectrophotometer. A reference cuvette lacking the chelator but

containing the antitumorigenic drugs was used as a reagent blank. To

169

achieve anaerobiosis all solutions were purged with argon, and glucose
and glucose oxidase were added to the incubation mixtures to remove
any residual oxygen. To prevent ferrous iron oxidation by H202,
catalase was also included. Incubation mixtures (1 ml final volume)
were constituted in 50 mM NaCl, pH 7.0 and reactions were started by
the addition of NADPH (0.5 mM) with other chemicals at concentrations
specified in Table and Figure legends. The amount of iron released
from ferritin was calculated using E - 12.11 mM’? cm’? for the
ferrouSfiggthgvphenanthroline complex. Measurement of the stimulation
of NADPH oxidation by the quinones was also monitored under anaerobic
conditions at 3A0 nm. Incubations (1 ml final volume) contained all
reagents at the same concentrations as indicated for the anaerobic
iron release assays except that NADPH was included at 0.2 mM.

ESR Measurements

 

Detection of the semiquinone free radicals of adriamycin,
daunomycin, and diaziquone was done using a Varian Century 112 ESR
spectrometer at room temperature under anaerobic conditions.
Spectrometer settings for incubations containing adriamycin were
3327.0 G magnetic field, 15 mw microwave power, 9.fl215 GHz frequency,
1000 KHz modulation frequency, 0.63 modulation amplitude, 0.5 sec time
constant, hx10” gain and a 100 G scan over A minutes. The conditions
which differed for daunomycin were that the magnetic field was 3329.2
G and the frequency was 9.u27u GHz. For detection of the diaziquone
radical the spectrometer settings which differed from that of the
anthracyclines were: 3320.0 G magnetic field, 9.H29H GHz frequency,
1.25 modulation amplitude, 0.25 sec time constant, and a 16 min time

scan 0

RESULTS

Aerobic and Anaerobic Release of Iron from Ferritin

As seen in Table 1, adriamycin and daunomycin, in the presence of
NADPH and NADPchytochrome PHSO reductase, were capable of releasing
iron from ferritin. However, SOD, a scavenger of 02?, caused only "81
and 311 inhibition of adriamycin and daunomycinvdependent iron
release, respectively, whereas SOD essentially completely inhibited
the release of iron from ferritin by the 027 generator xanthine
oxidase.

Table 1. Effect of SOD on Aerobic Release of Iron from

Ferritin by Xanthine Oxidase and the Redox Cycling of
Antitumorigenic Agents.

 

(nmol Fe2*/min)

 

System ~SOD +SOD
Xanthine oxidase 0.66 0.0”
NADPH'cytochrome Puso reductase 0.11 0.06
Plus: Adriamycin 0.50 0.26
Daunomycin 0.52 0.36
Diaziquone 0.12 0.04

 

Reaction mixtures (1 ml final volume) contained xanthine oxidase
(0.025 U) and xanthine (0.33 mM) or NADPchytochrome PMSO reductase
(0.10 U) and NADPH (0.50 mM), rat liver ferritin (1000 nmol Fe3*),
catalase (500 U), orthovphenanthroline (1 mM) and,'where indicated,
SOD (500 U), adriamycin (0.”0 mM), daunomycin (0.u0 mM), or diaziquone
(0.“0 mM) in 50 mM NaCl, pH 7.0. Reactions were continuously
monitored at 510 nm for formation of the phenanthrolinevferrous
complex.

170

171

The semiquinone radicals of several antitumorigenic compounds
including adriamycin have been shown to reduce ferric iron under
limiting oxygen concentrations (15) thus it was conceivable that the
SODvinsensitive rate of iron release from ferritin was due to direct
reduction of ferritin iron by the semiquinone radicals. Results shown
in Table 2 clearly demonstrate that under anaerobic conditions, where
the reaction of the semiquinone radical with oxygen is prevented, a
rapid rate of iron release from ferritin was observed.

Table 2. Anaerobic Release of Iron from Ferritin and NADPH Oxidation
by Antitumorigenic Agents.

 

iron released NADPH oxidation NADPH oxidized

 

 

System (nmol/min) (nmol/min) iron released

NADPR'cytochrome

PHSO reductase 0.00 0.12 ----

Plus: Adriamycin 3.05 11.57 3.79
Daunomycin 2.11 8.36 3.96
Diaziquone 0.00 6.11 ----

 

Reaction mixtures (1 ml final volume) contained NADPchytochrome P450
reductase (0.10 U),‘rat liver ferritin (25 nmol Fe3+), catalase (500
U), orthov phenanthroline (1 mM) and, where indicated, adriamycin
(0.25 mM), daunomycin (0.25 mM), or diaziquone (0.25 mM). Glucose (5
mM) and glucose oxidase (10 U) were included to assure anaerobiosis.
Reactions were initiated by the addition of NADPH (0.50 mM) and
continuously monitored at 510 nm. The stimulation of NADPH oxidation
by the antitumorigenic agents was also monitored anaerobically at 3&0
nm. Incubations were identical to those for measurements of anaerobic
iron release except that NADPH was 0.20 mM.

172

These data suggest that the semiquinone free radicals of adriamycin
and daunomycin rapidly release iron from ferritin. The ratio of the
rate of NADPH oxidation to the rate of iron release was nearly four
for both compounds indicating that the process was not tightly coupled
yet still resulted in a rate of iron release which was much greater
than that observed aerobically (Table 1).

The total amount of iron released increased as the concentration
of ferritin increased (Figure 1), although at all ferritin
concentrations, a maximum of 50-60% iron release was achieved in 15
minutes. Similar iron release kinetics and the resistance of a
portion of the iron to be released has been reported by others (16-18)
and is thought to be related to heterogeneity of the iron core.
However, this is still a much greater percentage of iron release than
was observed in the xanthine oxidase system where approximately 5% of
the iron within ferritin was released over 15 minutes (9). Similarly,
iron release was a function of the adriamycin concentration (Figure 2)
and the activity of NADPchytochrome Pu50 reductase (Figure 3).

If the cardiotoxicity resulting from anthracycline therapy is
indeed related to the release of iron from ferritin, it is important
to compare these compounds to one which is relatively nonvtoxic. We
chose to study diaziquone (also known as AZQ), an aziridinyl quinone
which is also reduced by NADPvaytochrome PHSO reductase to its
corresponding anion radical (19), but appears to be better tolerated
clinically (20). As shown in Tables 1 and 2, diaziquone was unable to
mediate the release of iron from ferritin either aerobically or
anaerobically. The diaziquone free radical autoxidizes much more

slowly than the anthracyclines (21,22). Perhaps explaining its

173

Figure 1. Effect of Varying Ferritin Concentration on the
Anaerobic Release of Iron from Ferritin by Adriamycin. Reaction
mixtures (1 ml final volume) contained NADPchytochrome PHSO
reductase (0.1 U), catalase (500 U), glucose (5 mM), glucose
oxidase (10 U), orthovphenanthroline (1 mM), adriamycin (0.2 mM),
and varying amounts of rat liver ferritin as indicated in 50 mM
NaCl, pH 7.0. Reactions were initiated by the addition of NADPH
(0.5 mM) and continuously monitored at 510 nm. 0 - no additions,
Aa 25 pH Fe3+, 0 - 50 11M Fe3+, o- 75 pl“! Fe3*,A- 100 pl“! M3+ as
ferritin. ‘

 

174

 

 

 

 

 

 

 

60"
5O -
g 40-
+ _ —l_‘
N
a:
ll.
30-
20-
+ A
IOP
O ‘ #VW 1 T
O 5 IO 15

TIME (min)

 

175

Figure 2. Effect of Varying Adriamycin Concentration on the
Anaerobic Release of Iron from Ferritin by Adriamycin. Reaction
mixtures (1 ml final volume) contained NADPchytochrome PHSO
reductase (0.1 U), catalase (500 U). glucose (5 mM). glucose
oxidase (10 U), orthovphenanthroline (1 mM), rat liver ferritin
(25 uM Fe3+), and varying concentrations of adriamycin as
indicated in 50 mM NaCl, pH 7.0. Reactions were initiated by the
addition of NADPH (0.5 mM) and continuously monitored at 510 nm.
0 - no adriamycin, A- .025 mM. I- 0.5 mM, 0 - .1 mM, and‘ D- .2
mM adriamycin. ' ‘ " ‘ '

 

 

 

176

 

T

 

 

 

 

 

‘1
“‘1
A
#9 o— —0
j I I
5 IO 15

TIME (min)

 

177

Figure 3. Effect of Varying NADPHvatochrome PVSO Reductase
Activity on the Anaerobic Release of Iron from Ferritin by
Adriamycin. Reaction mixtures (1 ml final volume) contained
adriamycin (0. 2 mM), catalase (500 U), glucose (5 mM). glucose
oxidase (10 U), ortho-phenanthroline (1 mM), rat liver ferritin
(25 pH Fe 3+), and varying activities of NADPchytochrome PHSO
reductase as indicated in 50 mM NaCl, pH 7. 0. Reactions were
initiated by the addition of NADPH (0.5 mM) and continuously
monitored at 510 nm. In no enzyme, 0 - .0125 units, 0- .025
units, A- .05'units, and 0 - .10 units of enzyme activity.‘

 

 

178

 

 

 

 

 

, l

18" d

15" ‘

A12- -

E

3

It.

a:

1.1.9.. ‘
5.. .
3. ..
O

O 5 -'0 15

TIME (min)

179

inability to stimulate iron release from ferritin aerobically.
Anaerobically, NADPH oxidation was stimulated by diaziquone,
suggesting that radical formation was occurring, but the radical was
apparently unable to mediate the release of iron from ferritin (Table
2).

Effect of Ferritin on ESR Detection of Free Radicals

To confirm and extend these results, ESR spectroscOpy was used to
detect the corresponding free radicals of the three drugs, and to
study their interaction with ferritin. As shown in Figures A, 5 and
6, spectra characteristic of the semiquinone radicals of adriamycin,
daunomycin, and diaziquone were obtained when the compounds were
incubated with NADPH-cytochrome PHSO reductase under anaerobic
conditions. The addition of 32 ug of ferritin (100 on Fe3+) resulted
in a marked diminution of the free radical signals of the
anthracyclines (Figures A and 5). However, ferritin had essentially
no effect on the free radical signal of diaziquone (Figure 6). Even
at higher ferritin concentrations the signal was not diminished
(results not shown). These results are indicative of electron
transfer from the semiquinone radicals of adriamycin and daunomycin to
ferritin with a concomitant loss of the free radical signals and a
release of ferrous iron.

As the reactions were scanned over time a gradual reappearance of
the free radical signals of adriamycin and daunomycin was observed as
the iron was released, allowing a detectable accumulation of the
radicals. Electron transfer may be directly to the ferric iron core
as has been pr0posed (17), or may occur at oxidovreduction sites on

the interior of the channels leading to the central core (23). This

18O

question has yet to be resolved for a number of reductants capable of
releasing iron from ferritin although the most recent evidence
indicates at least partial penetration of the shell by the reductant
is required (16). Irrespective of mechanism, ferritin did not quench
the free radical signal attributed to the aziridinyl semiquinone,
thereby confirming the inability of diaziquone to release iron from

ferritin.

 

 

181

. ..nco:uoz

can mHmHLoumz :« cocfiapao mm mcoapfiucoo mmm.nuH3 mao>fiuauoaot soccmon cam A29 m.ov moo<z
do eoqsaoos on» so oopssofiss osos osoaoosom .A+mom z: oops closeted dosed one new .AD o_v
ommoaxo omoosam .Aza my omoosam .Aza Fv cachamfigum .A: oomv ommampmo .A: F.0V composuog
omzm decenOOBhormma<z vocfimucoo nossao> Hanan as —v mmtsuxae cofiuomom .mmm he :oHpoouoo
Hmofiomm ococflsaaamm aaomsmaso< co cuuflsnmm no uoouum on» no snow m>HuHuoaom .n ossmam

182

q—

00.

m_

0.

35
ms: ._.

 

 

 

183

. ..nvocuoz

can namnnoumz an nocnauso mm aconunocoo mmm.nu«: nao>nuauoaon ooccmom one Axe m.ov mmo<z
no :oHpHoom on» no nonmagncn ego: macapomom .A+mon :3 oonv :Hannon Lo>HH own cum .As onv
ommunxo omoosam .Azs my omoosam .Aza FV anoneocsmo .A: oomv ommamumo .AD _.ov ommuozcon
om=m oeonnoounopmmo<z vocnmucoo Aoasao> Hanan as —V nonsuxna canuomom .mmw no :oHpoouoa
Hmoncmm ococnsaneom :Hozaocsmo co caunnnon no uoonnm on» no cmom o>nunuoaom .m onsmnm

184

_.||.|o|o_|||_

m.

0.

Es. .
MEI.

185

Figure 6. Repetitive Scan of the Effect of Ferritin on
Diaziquone Semiquinone Radical Detection by ESR. Reaction
mixtures (1 ml final volume) contained NADPchytochrome PHSO
reductase (O. 2 U), catalase (500 U), diaziquone (1 mM), glucose
(5 3mM), glucose oxidase (10 U), and rat liver ferritin (100 uM
Fe 3+). Reactions were initiated by the addition of NADPH (0. 5
mM) and scanned repetitively with ESR

 

 

 

 

 

TIME
(min)

10

I5

 

 

DISCUSSION

The demonstration of the release of iron from its storage protein
by the anthracycline antibiotics may be extremely significant with
respect to their toxicity as there exists an ever increasing body of
evidence which suggests that complexation with iron greatly
potentiates the cytotoxicity of adriamycin. Detailed studies have
demonstrated that adriamycin and daunomycin bind ferric iron at a
ratio of 3:1 with an overall association constant of 1033 (2H,25).

The adriamycinviron complex possesses a unique chemistry in that the
adriamycin is capable of reducing its bound iron (26). Subsequent
autoxidation of the ferrous iron apparently produces active oxygen
species (27) which may promote the oxidation of biological
macromolecules. Accordingly, adriamycinviron complexes have been
demonstrated to initiate lipid peroxidation via self'reduction (26) or
in the presence of reducing systems such as GSH or NADPH'cytochrome
Pfl50 reductase (2fl,28). Reduction of adriamycin by xanthine oxidase
or ferredoxin reductase has also been shown to damage DNA in an
ironvdependent fashion (29). Similarly, the adriamycinviron complex
was capable of cleaving DNA in the presence of H202 (30).

Thus, if oxidative stress is in fact responsible for the toxicity
of the anthracyclines it appears that, irrespective of mechanism, iron
is intimately involved. In agreement, the in 1119 cardiotoxicity of
adriamycin has been partially ameliorated by free radical scavengers
and iron chelators (31,32). Therefore, identification of physiologic
sources of iron is critical to an understanding of anthracycline

toxicity. Demant (33) has recently demonstrated in vitro a transfer

187

188

of iron from ferritin to adriamycin, dependent solely upon the high
affinity of adriamycin for iron, however, only 2% of the iron in
ferritin was released over 6 hours. Our present work demonstrates
that reduction of the anthracyclines by NADPchytochrome PHSO
reductase results in a much more efficient release of iron from
ferritin via two mechanisms. Redox cycling of the reduced drug
generates 027 which catalyzes a slow rate of iron release from
ferritin. Alternatively, the semiquinone free radicals of adriamycin
and daunomycin rapidly release much of the iron via reduction of the
ferric hydroxide core.

Heart tissue contains 30'60 ug of ferritin protein per gram (3“).
Ferritin generally averages 20% iron loading (8), therefore heart
tissue may contain up to 120 nmol iron per gram. In addition to
containing significant amounts of ferritin, cardiac cells also contain
only low amounts of catalase and SOD (35), that are important
antioxidant defense enzymes. The combination of iron availability
(which would promote the formation of damaging anthracycline-iron
complexes and subsequent oxygen radical generation) and low levels of
protective enzymes could perhaps explain the selective sensitivity of
the heart to anthracycline induced toxicity. Reduction of the
anthracyclines by nuclear and microsomal flavoproteins, and continued
redox cycling, would be expected to produce increasingly hypoxic
conditions within tissue, eventually mitigating conditions that favor
the rapid release of iron from ferritin. It is noteworthy that many
tumor cells contain high levels of ferritin (36) and that adriamycin
has been shown to be much more cytotoxic to hypoxic tumor cells (37),

conditions which this study suggests would favor the release of iron

189

from ferritin. Thus, it is possible that the selectivity of the
anthracyclines towards ne0plastic cells may also be related to an
interaction between ferritin and the drugs.

The inability of diaziquone, which appears to be well tolerated
in clinical studies (20), to release iron from ferritin may suggest a
potentially important means of diminishing the cardiotoxicity of
antitumor drugs. These studies do not allow us to distinguish whether
the aziridinyl structure is unable to gain access to the iron core, or
to oxidovreduction sites in the channels, or that the redox potential
of its radical (~168 mV) (22) is not sufficient to reduce ferritin
iron (-230 mV at pH 7.0) (38). However, these possibilities suggest
that altering the redox potential or sterically hindering access to
the inner core of ferritin to prevent iron release may be a
potentially effective means of diminishing cardiotoxicity.
Alternatively, administration of appropriate iron chelators such as
desferrioxamine may help to ameliorate anthracycline induced toxicity.
Alteration of structure to prevent iron binding also appears to
significantly lessen the ability of the compounds to potentiate
oxidative damage (7) and correspondingly, to lessen cardiac toxicity.
We suggest that the delocalization of tissue iron may be a common
feature contributing to anthracycline mediated toxicities. These
findings may aid in designing antidotes to diminish the cardiotoxicity
of existing drugs and in the development of new, less toxic

chemotherapeutic agents.

10.

11.

12.

LIST OF REFERENCES

N. B. Pratt and R. W Ruddon. The antibiotics. In: The
Anticancer Drugs, pp. 1M8v194. Oxford University Press, New
York (1979)

L. Lenaz and J.A. Page. Cardiotoxicity of adriamycin and
related anthracyclines. Cancer Treat. Rev. 3: 111'120 (1976).

R.A. Minow, R.S. Benjamin and J.A. Gottlieb. Adriamycin
(NSCv123127) Cardiomyopathy - An overview with determination
of risk factors. Cancer Chemother. Rep. 6: 195-210 (1975).

N.R. Bachur, S.L. Gordon, M.V. Gee and H. Kon.

NADPH cytochrome PvNSO reductase activation of quinone
anticancer agents to free radicals. Proceed. Natl. Acad. Sci.
76: 95N'957 (1979). ’ ' ' '

N. R. Bachur, M. V. Gee and R. D. Friedman. Nuclear catalyzed
antibiotic free radical formation. Cancer Res. 3g: 1078*1081
(1982). ' ‘ ‘ '

R.D. Olson, R.C. Boerth, J.G. Gerber and S. Nies. Mechanism
of adriamycin cardiotoxicity: evidence for oxidative stress.
Life Sci. 22: 1393'1u01 (1981). ‘

J. Muindi, B.K. Sinha, L. Gianni and C. Myers.

Thiolvdependent DNA damage produced by anthracyclineviron
complexes. The structure activity relationships and molecular
mechanisms. Molec. Pharmacol. 21: 356~365 (1985).

P. M. Harrison. Ferritin: An iron'storage molecule. Sem.
Hematol. 1": 55'70 (1977). '

C. E. Thomas, L. A. Morehouse and S. D. Aust. Ferritin and
superoxidevdependent lipid peroxidation. 'J. Biol. Chem. 260:
3275'3280 (1985) ' ’ ‘

R. F. Beers, Jr. and I. W. Sizer. A spectrophotometric method
for measuring the breakdown of hydrogen peroxide by catalase.
J. Biol. Chem. 195: 133~1u0 (1952).

J. M. McCord and I. Fridovich. Superoxide dismutase. An
enzymic function for erythrocuprein (hemocuprein). ‘J. Biol.
Chem. 2AM: 60u9v6055 (1969) ‘ '

T.C. Pederson, J.A. Buege and S.D. Aust. Microsomal electron
transport. The role of reduced nicotinamide adenine
dinucleotide phosphate'cytochrome 0 reductase in liver
microsomal lipid peroxidation. J. Biol. Chem. 238: 713Hv71u1
(1973). ° - ' ‘ ‘ -

190

13.

14.

15.

16.

17.

18.

19.

20.

21.

22.

23.

191

J.W. Halliday. Immunoassay of ferritin in plasma. In:
Methods in Enzymology, Volume 8” (J. Langone and J.
VanVunakis, eds ) DD. 1N8 171, Academic Press, New York
(1982) '

P.B. Brumby and V. Massey. Determination of nonheme iron,
total iron, and copper. In: Methods in Enzymology, Volume 10
(R. N. Estabrook and M. E. Pullman, eds.) pp. 163-u7u, Academic
Press, New York (1967).

J. Butler, B.M. Hoey and A.J. Swallow. Reactions of the
semiquinone free radicals of'antivtumor agents with oxygen and
iron complexes. FEBS Lett. 182: 95'98 (1985).

F. Funk, J. P. Lenders, R. R. Crichton and W. Schneider.
Reductive mobilization of ferritin iron. 'Eur. J. Biochem.
152: 167 172 (1985). ' ' '

T. Jones, R. Spencer and C. Walsh. Mechanism and kinetics of
iron release from ferritin by dihydroflavins and dihydroflavin
analogues. Biochemistry 11: uo11vuo17 (1978).

T.C. Roy, P.M. Harrison, M. Shabbir and J.G. Macara. The
release of iron from horse spleen ferritin t0” ‘
1,10~phenanthroline. Biochem. J. 137: 67'70 (197”).

P.L. Gutierrez and N.R. Bachur. Free radicals in quinone
containing antitumor agents. The nature of the diaziquone
(3,6vdiaziridinyl~2,5vbis (carboethoxyamino)~1,H'benzoquinone)
free radical. Biochim. Biophys. Acta lgg: 37*“1 (1983).

G.A. Curt, J.A. Kelley, C.V. Kufta, B.H. Smith, P.L.
Kornblith, R.C. Young and J.M. Collins.' Phase II'and
pharmacokinetic study of aziridnylbenzoquinone

[2, 5*diaziridinylv3, 6vbis (carboethoxyamino)~1, uvbenzoquinone,
diaziquone. NSC 182986] in high grade gliomas. Cancer Res.
N3: 6102-6105 (1983) '

P. L. Gutierrez, M. J. Egorin, B. M. Fox, R. Friedman and N. R.
Bachur. Cellular activation of diaziquone

[2, 5-diaziridinyl-3, 6~bis (carboethoxyamino)~1,flvbenzoquinone]
to its free radical species. Biochem. Pharmacol. i3:
1uu9v1uss. ' '

B.A. Svingen and G. Powis. Pulse radiolysis studies of
antitumor quinones: radical lifetimes, reactivity with
oxygen, and onevelectron reduction potentials. Arch. Biochem.
Biophys. 209:119 126 (1981) ' '

C.F.A. Bryce and R.R. Crichton. The catalytic activity of
horse spleen apoferritin. Preliminary kinetic studies and the
effect of chemical modification. Biochem. J. 133: 301'309
(1973). ' ' ' '

2H.

25.

26.

27.

28.

29.

30.

31.

32.

33.

3".

35.

192

P. M. May, G. N. Williams and D. R. Williams. Solution chemistry
studies of adriamycin-iron complexes present in vivo. Eur. J.
Cancer 16: 1275 1276 (1980)

C.E. Myers, L. Gianni, C.B. Simone, R. Klecker and R. Greene.
Oxidative destruction of erythrocyte ghost membranes catalyzed
by the doxorubicinviron complex. Biochem. 31: 1707*1713
(1982). ‘ ‘ ‘ ' '

K. Sugioka and M. Nakano. Mechanism of phospholipid
peroxidation induced by ferric ionvADPvadriamycin
co-ordination complex. Biochim. Biophys. Acta 713: 333v3h3
(1982). ' ' ‘

S. D. Aust, L. A. Morehouse and C.E. Thomas. Role of metals in
oxygen radical reactions. J. Free Rad. Biol. Med. 1: 3'25
(1985). ' - - ' - °

E.G. Mimnaugh, T.E. Cram and M.A. Trush. Stimulation of mouse
heart and liver microsomal lipid peroxidation by anthracycline
anticancer drugs: Characterization and effects of reactive
oxygen scavengers. J. Pharmacol. Exper. Ther. 339‘ 806-813
(1983). ' ' ' ' ‘

D.A. Rowley and B. Halliwell. DNA damage by
superoxidevgenerating systems in relation to the mechanism of
action of the antivtumor antibiotic adriamycin. Biochim.
Biophys. Acta 761: 86'93 (1983). ‘

H. Eliot, L. Gianni and C. Myers. Oxidative destruction of
DNA by the adriamycin iron complex. Biochem. 2;: 928'936
(198A). '

C.E. Myers, H.P. McGuire, R.R. Liss, I. Ifrim, K. Grotzinger

and R. 0. Young.“ Adriamycin: The role of lipid peroxidation

in cardiac toxicity and tumor response. Science 197: 165-167
(1977). ‘ ' '

E.R. Herman, A.M. ElvHage, V.J. Ferrans and D.T. Hitiak.
Reduction by ICRFv187 of acute daunorubicin toxicity in Syrian
golden hamsters. Res. Comm. Chem. Pathol. Pharmacol. fig:
217*231 (1983). ‘ ' ‘ ‘

E. J. F. Demant. Transfer of ferritinvbound iron to adriamycin.
FEBS Lett. 176: 97-100 (198M).

A. Bezkorovainy. Chemistry and biology of iron storage. In:
Biochemistry of Nonheme Iron (E. Frieden, ed.) pp. 207v269,
Plenum Press, New York (1980).

 

J.M. Doroshow, G.Y. Locker and C.E. Myers. Enzymatic defenses
of the mouse heart against reductive oxygen metabolites.
Alterations produced by doxorubicin. J. Clin. Invest. 62:
128v135 (1980). ‘ ‘ ‘

36.

37.

38.

193

C. Cohen, G. Shulman and L.R. Budgeon. Immunohistochemical
ferritin in testicular seminoma. Cancer fig: 2190*219“ (198A).

B.A. Teicher, J.S. Lazo and A.C. Sartorelli. Classification
of antivneoplastic agents by their selective toxicities toward

oxygenated and hypoxic tumor cells. Cancer Res. 31: 73'81
(1981). ‘ ‘ '

G.D. Watt, R.B. Frankel and G.C. Papaefthymiou. Reduction of
mammalian ferritin. Proceed. Natl. Acad. Sci.Ԥg: 36H0v36u3
(1985). ‘ ' ‘ ' '

CHAPTER V

RAT LIVER MICROSOMAL NADPH'DEPENDENT RELEASE OF IRON FROM
FERRITIN AND LIPID PEROXIDATION

194

ABSTRACT

Microsomes prepared by the usual method of differential
centrifugation were found to contain ferritin, SOD, and catalase which
could be separated from microsomes by chromatography on Sepharose
CLvZB. Addition of purified rat liver ferritin to chromatographed
microsomes resulted in a significant stimulation of NADPHvdependent
lipid peroxidation which was inhibited by exogenously added SOD. Iron
release from ferritin by these microsomes was also inhibited by SOD.
Ferritin did not promote NADPHvdependent microsomal lipid peroxidation
when added to microsomes isolated in the usual manner, presumably due
to the endogenous SOD present in the microsomes. Accordingly, only
very low rates of iron release from ferritin were observed with these
microsomes. Paraquat,which generates 027 via redox cycling, greatly
stimulated iron release from ferritin and lipid peroxidation in
chromatographed microsomes. Paraquat had no effect on iron release
from ferritin or lipid peroxidation in microsomes which were not
chromatographed unless they were first treated with CN' to inhibit
endogenous SOD. These studies indicate that the majority of
microsomal iron is contained within ferritin and that following
release by 027.this iron serves to promote the peroxidation of

microsomal lipids.

195

INTRODUCTION

The ability of iron to promote microsomal lipid peroxidation was
first reported by Hochstein gt El: (1) and numerous studies have since'
confirmed the integral role transition metals play in the peroxidative
process (Z-A). Most in vitro studies have employed low molecular
weight iron complexes such as ADPvFe3+ because they are envisioned to
exist within the cell. Some evidence for such complexes in
reticulocytes has been presented (5,6), however an analagous iron
chelate has not been conclusively identified in other tissues (7).

The majority of cellular iron is stored within ferritin as a
ferric hydroxide core complexed with phosphate (8). Some time ago it
was proposed that much of the non-heme iron present in microsomes may
be ferritin but it was suggested that this iron was unavailable for
peroxidation (9). Ferritin was later identified as covsedimenting
with the microsomal fraction (10) and others (11) further demonstrated
that a portion of the ferritin was tightly associated with the
microsomal membranes. However, it is unlikely that this ferritin iron
is available to generate an oxidizing species as it is surrounded by
the spherical protein shell. Mobilization of iron from ferritin
appears to require reduction and is enhanced by chelators (12),
conditions that can also promote lipid peroxidation. Accordingly,
Wills (13) and Gutteridge (1“) have demonstrated that ferritin iron is

released in an ascorbate-dependent lipid peroxidation system.

196

197

Recently, Rowley and Sweeney (15) have demonstrated that
NADPH-cytochrome PA50 reductase (cytochrome c reductase) is capable of
releasing ferritin iron in the presence of FMN under anaerobic
conditions. Similar results have previously been reported for
mitochondria that also appear to contain distinct binding sites for
ferritin (16,17). These reports indicate that ferritin may not serve
only as an iron storage protein but may play a more dynamic role in
the cellular metabolism of iron. We have recently demonstrated that
027 can release iron from ferritin (18,19) in agreement with others
(20). It has also been shown that microsomes generate small amounts
of 027 (20'22) and that SOD inhibits microsomal lipid peroxidation
(21). These results, in conjunction with the apparent association of
ferritin with microsomal membranes, have led us to investigate whether
iron can be released from ferritin by microsomes aerobically and

subsequently promote the peroxidation of microsomal lipids.

MATERIALS AND METHODS

Materials

NADPH, ADP, cytochrome 0 (Type VI), Z-TBA, H,7~diphenyl~1,10v
phenanthroline, bathophenanthroline sulfonate, paraquat,
‘ggthg-phenylenediamine, pagavnitrophenylvaacetylvB-Dvglucosaminide,
and butylated hydroxytoluene were from Sigma Chemical Company (St.
Louis, MO). Thioglycolate and sodium hydrosulfite were purchased from
Fischer (Fairlawn, NJ), sodium cyanide from Baker Chemical Company
(Phillipsburg, NJ) and H202 from Mallinckrodt Chemical Works (Paris,
KY). Imferon was a gift from Merrell Dow Pharmaceuticals (Cincinnati,
OH). All buffers and reagents were passed through Chelex 100 (BIO’Rad
Laboratories, Richmond, CA) ion exchange resin to free them of
contaminating transition metals.
Enzymes

Bovine erythrocyte SOD (EC 1.15.1) was obtained from Sigma
Chemical Company and catalase (EC 1.11.1.6) from Millipore (Freehold,
NJ). Catalase was chromatographed on Sephadex Gv25 (Pharmacia Fine
Chemicals, Piscataway, NJ) prior to use to remove the antioxidant
thymol. Superoxide dismutase activity was measured by a modified
method of McCord and Fridovich (23) using acetylated cytochrome c
prepared as per Morehouse 35 El- (22). Catalase was assayed according
to Beers and Sizer (2“) while NADPchytochrome PHSO reductase activity
was determined by cytochrome 0 reduction (25) and cytochrome PA50 from
its carbon monoxide difference spectrum (26). Hexosaminidase activity

was assayed as described by Horvat gt El: (27).

198

199
Preparation of Antibodies and Ferritin Quantitation

Ferritin was purified from the livers of rats given an
intraperitoneal injection of Imferon (ironvdextran) as described by
Halliday (28) with modification (18). The purified protein was mixed
1:1 with Freund's adjuvant and injected subcutaneously (1 ml total
volume containing 1 mg protein) into the backs of rabbits. Following
three biweekly injections the animals were bled from the marginal ear
vein. The serum obtained was precipitated with ammonium sulfate and
chromatographed on DEAR-cellulose to obtain a purified IgG fraction.
Purity was determined by polyacrylamide gel electrophoresis.

Ferritin in microsomes was quantitated using a modification of an
ELISA previously described (29). Microtiter plates (Dynatech 96 well,
Alexandria, VA) were coated with 5 us (200 pl total volume) of
purified rat liver ferritin in 50 mM sodium bicarbonate buffer, pH
9.6, at HOG for 15 hours. The plates were rinsed with water and
coated with 0.1% gelatin in PBS, pH 7.5. After incubation at 37°C for
30 min the plates were rinsed with water and 3 ug of anti-ferritin IgG
plus microsomes (various dilutions containing 1.15 r 5 us protein), or
purified ferritin (0'50 ng) for the standard curve, in 0.21 gelatin in
PBS containing 0.2% Tween 20 (total volume 100 pl). The plates were
then carefully washed with water after a one hour incubation at 37°C
and 50 ul of a 1:2000 dilution of goat, antivrabbit horseradish
peroxidase coupled IgG (Cappel Laboratories, Cochranville, PA) was
added in PBS containing 11 gelatin and 0.1% Tween 20 and allowed to
react for 30 min at 37°C. The plates were again thoroughly rinsed and

substrate, 2.2 mM ortho-phenylenediamine in 0.1 M citrate, pH 5.0,

200

containing 2.6 mM H202 (total volume 100 pl), was added. The reaction
was stopped after 10 min by the addition of 50 pl of A N H2803 and the
absorbance recorded at "90 nm.
Preparation of Microsomes

Male SpraguevDawley rats (250-275 g) were obtained from Charles
River (Boston, MA) from which liver microsomes were isolated as per
Pederson and Aust (30) with the exception that an additional
centrifugation at 25,000 x g prior to ultracentrifugation was included
to ensure removal of lysosomes. The pellet obtained after
centrifugation at 105,000 x g was rehomogenized in 0.02 M
Tris~HCl/O.15 M KCl pH 7.A and applied to a Sepharose CLvZB column
(2.5 x 25 cm) equilibrated in the same buffer (31). Microsomes, which
eluted in the void volume, were pooled and centrifuged again at
105,000 x g and resuspended in 50 mM NaCl containing 501 glycerol.
All solutions utilized were thoroughly purged with argon and all steps
performed at AOC to minimize autoxidation of unsaturated lipids.

Lipid Peroxidation Assays

 

NADPHvdependent peroxidation of microsomes was performed by
incubating microsomes (0.5 mg/ml) with NADPH and other additions as
specified in the figure legends. Reaction mixtures were constituted
in 50 mM NaCl, pH 7.0, and incubated at 37°C in a Dubnoff metabolic
shaker under an air atmosphere. Although unbuffered, incubations
remained at pH 7.0 throughout the course of the experiments.
Peroxidation was monitored by taking aliquots from the incubations at
0, 8, 16, and 2“ min to measure the rate of MDA formation using the

TBA test (32). Rates shown are those calculated at 16 min.

201

Assays for Total Iron and Ferritin Iron Release

Total iron was determined by the method of Brumby and Massey (33).
The release of iron from ferritin was measured according to Ulvik and
Romslo (16) with modification. Microsomes (2 mg/ml) were incubated in
an Open cuvette in 50 mM NaCl, pH 7.0, containing 175 uM
bathophenanthroline sulfonate and catalase (1000 U/ml) to prevent
ferrous iron oxidation (18), with other additions as indicated in the
figure legends. The formation of the ferrous-bathophenan-
throline complex was determined by continuously monitoring the
difference in absorbance between 530 and 560 nm using the dual
wavelength, non-scan mode of an Aminco DW-2 UV/VIS spectrophotometer
(16). Reactions were started by the addition of NADPH and the amount
of iron released from ferritin was determined from a standard curve

using ferric chloride reduced with 0.1 ml of 10% thioglycolate.

RESULTS

Chromatography of Microsomes and Activities of Associated Enzymes

Protein elution profiles obtained from a Sepharose CLvZB column
following chromatography of microsomes and purified rat liver
ferritin, which were run separately, are shown in Figure 1A. When
microsomes were chromatographed a readily visible yellow peak trailed
the microsomal fraction (identified by NADPH-cytochrome PNSO reductase
activity) and was found to elute at essentially the same volume as did
purified ferritin. This fraction was found to react with
antivferritin antibody using Ouchterlony double diffusion analysis
(Figure 1B). The ferritin content of the microsomes, determined using
an indirect competitive ELISA, was 6.97 NE of ferritin per mg of
microsomal protein, however following chromatography little ferritin
could be detected (Figure 1B and Table 1). This suggests that
ferritin was effectively removed by chromatography as the ELISA was
able to detect very low levels of ferritin as indicated in Figure 2.
In agreement, the total iron concentration of the microsomes was
greatly decreased by chromatography. Microsomes contained low
hexosaminidase activity (3.5 nmol pvnitrophenollmin/mg protein)
therefore the ferritin was unlikely a result of lysosomal
contamination.

The activities of NADPchytochrome PA50 reductase, cytochrome
PNSO, SOD, and catalase were measured before and after chromatography
to assess the effects of chromatography on the microsomes. Table 1
demonstrates that the specific activity of NADPchytochrome PN50
reductase and the specific content of cytochrome Pu50 are increased by

chromatography, a likely result of the removal of loosely associated
202

203

. .cuufipnmu

.85: pm; 8:23 o .mmaomopoae nl .mfimzmcm cot; an 5358.“ EB 3333.. $3262
omzm oeoncoou501zao<z >9 uofiufipcmofi mmeomOEan new a: 0mm um mocmnnomnm up consamcm

moo: mcofiuomam .mcocpoz cam mamfigmumz cu omcHHpso mm.acssaoo So mm x m.mv mmiqo omoamcqmm
co umcnmnmoumeocno mampmpmamm mam: :.> ma .H02 2 mP.oxaum1maLe 2 No.0 ca caufipuou Lm>HH
um; cmfiufinsq to Ama oo: zamumsfixogaamv mosomogon .mm1qu mmogmsaom co cHuHLme Lm>uq

pmm coauagsm cam mmsomonofiz no xcamgmoumaogno on» Lou oafluopm coHuSHm :Hmuonm .49 mpsmfim

204

 

 

‘15“
22

\-
24

20'

18

IO

 

 

 

|.5

o “2
-'- O
(uounup Ole-nos 2 '0'0

0

Fraction No.

205

Figure 1B. Ouchterlony Double Diffusion Analysis of Microsomes.
The center well contained antivferritin IgG (0.3 mg). Other
wells contained: A, microsomes (0.5 mg); B, microsomes (2.5 mg);
C, chromatographed microsomes (0.75 mg); D, yellow peak from
Sepharose CL'ZB column (fractions 1A~18 0.08 m8); E, purified rat
liver ferritin (0.08 mg); F, previmmune IgG (0.3 mg).

206

 

207

.nuonpmn cam mandamumn

:H vonunommo «qum m>HuHumaaoo .uomgquca on» an oozmmmm mm: nods: Ame omiov :Huaggmu

Lm>da pm; coaufipsq means omchuno mm: m>cso unmeasum one

.o>aso otmoeoom «mHam .m oasmaa

208

 

 

 

50

40

IO

 

l l l I
«2 :0 at «.1 0
O

O

0 0°
(“1” 0617) V

119 FERRITIN

209

.h I c .Oh Hmuou

com .Anomo massacm o— cv mcofiumgmaoaq Hmsomogofle oumgmqmn moss» I : cams: .o. m + coo: m

pmuonuox.ocm mamagmumx cu confidence mm cozmmmm one:
nosmuce oopmdoommm cam .coLH Hobo» .cauflssmm .moonumz ucm mamfigoum: ca omcfiauso mm mm1qo
encamnaom co anamLmOBmBOLco Loxccm coHummsuHLucmo Hmfipcmgmuuwu an omcmqoga mom: mmeomOLoH:

 

2.08:5 o.m.+.o.m new; Féumfi 8.0“.86

moaomonoaz

 

 

8.9.8.0

cocamchpmsoLno

RENEE 063.? E.J....Sm m.mm.+.\..mm_ 2.03.0.0 6.38.0 888.82
ma\mn ms\aoec ms\: max: ma\aos: ms\:
:HuHLLom om Hmpoe mom encampmo omzm composomm omza
oeonnoouzo msonnoouho

 

momsomonofiz spa: uoumwoomm< nosancm cam .coLH

Hmuop .cauficcmm co unamumoumsonno mmiqo omonmnnom ho aomuum

.F mHQMH

210

protein from microsomes. Catalase activity in microsomes was quite
high, in agreement with others (3fl.35), however, much of that activity
was removed during chromatography. Superoxide dismutase activity,
measured as the inhibition of xanthine oxidase-dependent reduction of
cytochrome c (23), was also detected in microsomes. Acetylated
cytochrome c was used as it is not directly reduced by NADPHv
cytochrome PASO reductase and is therefore more specific for
Oztvdependent reduction. Again most of the SOD activity was removed
by the chromatography step. These results were also confirmed by
subjecting microsomes and chromatographed microsomes to polyacrylamide
gel electrophoresis.under nondenaturing conditions and staining the
gel for SOD activity using the procedure reported by Beauchamp and
Fridovich (36) (data not shown).

Ferritin Iron Release

 

Since it was previously demonstrated that 027 could release iron
from ferritin (13-20) and since microsomes are reported to generate
027, it was of interest to determine first whether microsomes could
release ferritin iron and secondly, whether 02? production was
necessary for iron release. Because microsomes contain SOD these
experiments were conducted with chromatographed microsomes. As shown
in Figure 3 the rate of NADPHvdependent iron release from rat liver
ferritin was 0.1“ nmol Fe2+ released per min and could be increased
about 7vfold by the addition of 0.5 mM paraquat. Figure A
demonstrates that rates of iron release were essentially linear with
respect to microsomal protein at concentrations greater than 0.25 mg
protein per ml. In the absence,of microsomes less than 0.03 nmol Fe2+

per min was released from ferritin. Iron release was also linear with

211

Figure 3. Effect of Varying Paraquat Concentration on
NADPHvDependent Iron Release from Ferritin. Incubations (1 ml)
contained chromatographed microsomes (2 mg), ferritin (1 mM
Fe3*), NADPH (0.5 mM), catalase (1000 U/ml), bathophenanthroline
sulfonate (175 uM), and paraquat at the concentrations indicated
in 50 mM NaCl, pH 7.0. Formation of the
ferrousvbathophenanthroline complex was continuously monitored at
530 nm as described under Materials and Methods.

212

 

l.2 "

2+ .
nmol Fe released /m1n

 

 

o - I I 1 L 1

0 0.10 0.20 0.30 0.40 0.50
Paraquat (mM)

 

213

.moonumz cam mamasmpmz Loon: omcaauso

mm xmaaeoo on“Hognpcmcmcaonumn1msoggou 0:» ho cowumsnou Lou uoaouficoa mamsoscaucoo

. one: mcofiuommm .o.~ ma .Homz :3 om.:« A2: mprv oumcouasm ocaaognpcmcmnaonumn cam
Axe m.ov oncomeea .Aaa\= ooopv onnflmomo .Aza m.ov mao<z .A+moa 2a av saucepan .ooomoaoafi
mm mosomosoHs occamgmoumeonno omcamucoo has Fv macapmnsocH .CHawLme scan mommamm

cocH pcoocoamaimao<z co coaumgucmocoo cfimuosa Hmsomosofiz mzfixsm> no uoouum .z madman

211+

 

3.5 5038 32382.2
o.~ 0.. o._ no o

III-l d u a O

I
1"!
O

I
¢.
0

I I

a; m

0 O
nun/pascaleiflag lowu

 

 

 

215

respect to ferritin concentration over the range tested (Figure 5).
Paraquat was included in both of these experiments to ensure that 02:
was not rate limiting. In the absence of ferritin a small amount of
iron release was observed, however this rate subsided within 5 min.
This iron release probably represents the mobilization of some as yet
unidentified nonvheme, nonvferritin iron from microsomes.

A comparison of iron release from ferritin by microsomes and
chromatographed microsomes was also made (Table 2). Rates of iron
mobilized are normalized to 1 unit of NADPchytochrome PHSO reductase

activity to account for the different specific activities of the

 

enzyme in the two microsomal preparations. With microsomes, the rate
of iron release was only slightly increased upon the addition of
purified ferritin and addition of paraquat or SOD had little effect.
When these microsomes were treated with 2 mM CN', which is known to
inhibit the Cu, Zn~SOD, a 50% stimulation in the rate of iron release
was observed upon the addition of ferritin. Accordingly, in the
presence of CN’, paraquat now greatly stimulated iron release from
ferritin. When ferritin was added to chromatographed microsomes,
which are essentially free of SOD activity, iron release was increased
over 2vf01d, which could be inhibited by the addition of exogenous SOD.
Paraquat greatly stimulated iron release irrespective of the presence
of CN'. In the presence of CN’, paraquat, and ferritin, rates of iron
release were comparable for the two microsomal preparations.
NADPHvDependent Lipid Peroxidation Using Purified Ferritin

The demonstration that microsomes could catalyze the
NADPHvdependent release of iron from ferritin suggested that ferritin

may provide a source of iron for promotion of lipid peroxidation. In

216

Figure 5. Effect of Varying Ferritin Concentration on
NADPHvDependent Iron Release from Ferritin. Incubations (1 ml)
contained chromatographed microsomes (2 mg), NADPH (0.5 mM),
paraquat (0.5 mM), catalase (1000 U/ml), bathophenanthroline
sulfonate (175 pH), and varying concentrations of ferritin in 50
mM NaCl, pH 7.0. Formation of the ferrousvbathophenanthroline
complex was continuously monitored as described under Materials
and Methods.

released/min

2+

nmolFe

L2

217

 

 

O

I
200

4 I 4
400 600 800
Ferritin (uM Fe)

 

I
1000

 

J

218

Table 2. Effect of SOD, CN', and Paraquat on Ferritin Iron
Release by Microsomes

 

nmol Fe2* released/min/U reductase

 

Chromatographed
Microsomes Microsomes

' - NADPH 0.00 0.00
- Ferritin 0.78 0.12
Complete 0.88 0.91

+ SOD 0.88 0.62
+ CN' 1.28 0.83
+ paraquat 1.17 3.75
+ paraquat, + SOD 1.17 0.67
+ paraquat, + CN' 3.89 3.67

 

Incubations (1 ml) contained microsomes (2 mg for both preparations),
catalase (1000 U/ml) and bathophenanthroline sulfonate (175 u“) in 50
mM NaCl, pH 7.0. Where indicated additions were as follows: NADPH
(0.5 mM), ferritin (1 mM Fe3+), 300 (100 U/ml), CN’ (2 mM), and
paraquat (0.5 mM). Formation of the ferrous-bathophenan—

throline complex was determined by continuously monitoring the
difference in absorbance between 530 and 560 nm using the dual
wavelength, nonvscan mode of an Aminco DW~2 UV/VIS spectrophotometer
(Results are the averages of two separate experiments).

219

the presence of NADPH, microsomes isolated by conventional
differential centrifugation were found to exhibit greater rates of MDA

formation than chromatographed microsomes (Table 3).

Table 3. Effect of Catalase, SOD and Paraquat on NADPH-
Dependent Microsomal Lipid Peroxidation.

 

nmol MDA/min/U reductase

 

No Additions + Paraquat + Catalase + SOD

Microsomes 3.75 3.A7 “.03 3.75

+ Ferritin 3.75 2.67 “.25 “.25
Chromatographed

Microsomes 1.37 1.56 1.37 1.37

+ Ferritin ”.38 5.7“ ”.30 1.75

 

Reaction mixtures (2.5 ml, final volume) contained microsomes (0.5
mg/ml) and ADP (100 u”) in 50 mM NaCl, pH 7.0. Where indicated
additions were as follows: ferritin (200 uM Fe3+), paraquat (0.5 mM),
catalase (100 U/ml) and SOD (100 U/ml). Reactions were initiated by
the addition of NADPH (0.5 mM) and aliquots from the reaction mixtures
were assayed as described in Materials and Methods to determine the
rate of MDA formation. (Results are the averages of two separate
experiments). '

Again, rates are expressed per unit of NADPH'cytochrome PHSO reductase
activity to permit direct comparisons. However, upon the addition of
purified rat liver ferritin to lipid peroxidation incubations an
increase in MDA formation was noted only in chromatographed microsomes.
The addition of exogenous SOD prevented the ferritinvdependent

increase in peroxidation of chromatographed microsomes but had no

effect on the rate observed in the absence of ferritin for either

220

microsomal preparation, or the rate observed with microsomes plus
ferritin. The inclusion of paraquat resulted in a marked stimulation
of MDA formation in chromatographed microsomes but only when ferritin
was included. The addition of CN’ or DDC to inhibit endogenous SOD
activity in microsomes resulted in a nonvspecific inhibition of
NADPHvdependent lipid peroxidation perhaps either by inhibition of
microsomal monoxygenase activity (38) or by free radical quenching r
(39). Therefore, lipid peroxidation experiments analagous to the iron
release studies with microsomes (Table 2) are not shown.

It was of interest to determine if a difference in the amount of

 

iron available for the formation of an initiator was responsible for
the greater rates of lipid peroxidation in microsomes than
chromatographed microsomes. To assess this, NADPHvdependent lipid
peroxidation incubations were titrated with EDTA, which will inhibit
lipid peroxidation when it is in excess of iron (37). As shown in
Figure 6, peroxidation in microsomes was significantly inhibited by
approximately A uM EDTA. Lipid peroxidation in chromatographed

microsomes was similarly inhibited by approximately 2 uM EDTA.

221

 

. .moonuo:

. cam mamnpoumz an nonnnomoc mm condoms one: monsuxqa conuomon on» eonn.muo:UHHm new
“:5 m.ov mma<z no conpnoom 0:» no coumHuHcH,osoz meonpommm .o.» ma .Homz :5 om an «new
no mucsosm mcnnnm> cam Aas\ms o.—n moeomonoHs oochpcoo Amasao> Hmcnn .Ha m.mv.mmnsuxna
conuomom .conpmonxogmm manna Hmsomononz ucmocoaoaimao<z no conumgunw <eom .o onsmnn

222

223 arrow

 

 

 

mod

(0
Q
0

0:
Q
o .
|'-l‘/“!u"/V(1|I‘Jl0“-Iu

N_.O

 

9.0

 

DISCUSSION

While a requirement for transition metals including iron for
promotion of lipid peroxidation has been demonstrated in vitro (1-A)
the nature of the iron complex(es) able to catalyze redox reactions in_
1119 remains unknown. Most in vitro microsomal lipid peroxidation
studies have utilized exogenously-added low molecular weight ferric
iron chelates as the iron source. However, our results have confirmed
original studies (9) demonstrating that microsomes contain measurable
quantities of endogenous iron which can promote lipid peroxidation.
Extreme care was taken to remove metals from all solutions thus it
appears that this iron is associated with the microsomal fraction and
is not due to the contaminating iron often found in buffers, etc.

The present study has also confirmed previous reports (10,11)
that much of the endogenous iron in microsomes isolated by the usual
method of differential centrifugation is ferritin. Microsomes
prepared by differential centrifugation were found to contain 6.97 us
of ferritin per mg of protein. Ferritin (Mr - “A0 kdaltons) can
contain up to “500 atoms of iron per molecule but generally averages
20: loading (8). Thus, using these estimates, the iron content of
ferritin is 2.0M nmol Fe per ug of ferritin. Therefore the iron in
ferritin accounts for 1A.2 nmol Fe per mg of microsomal protein or,
subtracting 1.5 mg of the total iron per mg protein (Table 1) for
cytochromes PHSO and b5, 83% (1A.2/17.1) of the nonvheme iron in

microsomes. As microsomes have been shown to produce 02? (21,22), and

it has recently been demonstrated that 027 can reductively release

iron from ferritin (18'20), the possibility existed that the
223

 

224

endogenous iron promoting microsomal lipid peroxidation may have
originated from ferritin. To demonstrate this it was necessary to
first remove endogenous ferritin from microsomes. Treatment of
microsomes with calcium or antibody precipitation has been shown to
remove a portion of the ferritin (10), however, we found that
chromatography of microsomes on Sepharose CLvZB was a more rapid,
effective means for separating ferritin from the microsomes. ET
The addition of purified rat liver ferritin to chromatographed %
microsomes resulted in a significant stimulation of lipid peroxidation

which could be completely reversed by exogenous SOD. In agreement,

 

when paraquat was included to increase 02? production rates of MDA
formation in the presence of ferritin were greatly enhanced. Iron
release studies confirmed that chromatographed microsomes can catalyze
NADPHvdependent release of ferritin iron under aerobic conditions.
Iron release from ferritin was stimulated by paraquat and inhibited by
SOD, further indicating that release is Oztvdependent, in agreement
with the lipid peroxidation data.

When purified ferritin was added to microsomes isolated by
differential centrifugation there was no stimulation of
NADPHvdependent lipid peroxidation. These results, in conjunction
with the inhibition of ferritinvdependent lipid peroxidation by SOD in
chromatographed microsomes suggested that microsomes prepared using
differential centrifugation may contain SOD. Assay of microsomes for
SOD activity by inhibition of acetylated cytochrome c reduction or by
activity staining in polyacrylamide gels (results not shown) revealed

that this was indeed the case. The SOD present was apparently the

225

cytosolic Cu, Zn form as its activity was inhibitable by CN'.
Following chromatography, microsomes were essentially free of
endogenous SOD activity when measured with either method.

Again, iron release experiments confirmed the results obtained in
the lipid peroxidation experiments. When ferritin was added to the
microsomes essentially no increase in iron release was observed.
However, when the microsomes were pretreated with CN’ iron release
from ferritin was observed and paraquat also greatly stimulated iron
release. When both microsomal preparations were treated with CN’, and
paraquat was included, rates of iron release from ferritin were
virtually identical.

Interestingly, in the absence of added ferritin, microsomes
exhibited greater rates of peroxidation than chromatographed
microsomes. To ensure that the mechanism of iron release from
endogenous ferritin was the same as that for exogenously added,
purified ferritin, the partially pure ferritin fraction from the
Sepharose CLvZB column (fractions 1hv18, Figure 1A) was collected.
Addition of this fraction (200 uM Fe3*) to peroxidation experiments
using chromatographed microsomes gave results similar to those in
Table 3 in the presence and absence of paraquat (not shown),
indicating that 027 was also required to release the iron from the
endogenous ferritin associated with microsomes. Thus, since it
appeared that iron release from endogenous ferritin was unlikely to
account for the higher rates of peroxidation in microsomes (due to the
presence of SOD), another form of iron in these microsomes must be
responsible for promoting lipid peroxidation. These results are in
agreement with Montgomery §t_al. (10) who postulated that microsomes

contain a small, nonvheme iron pool which is not ferritin.

 

226

Attempts to further identify and quantitate this nonvferritin
iron pool were not successful. However, by titrating lipid
peroxidation incubations with EDTA, the size of the iron pool was
estimated to be 3vfl nmol iron per mg protein in microsomes and 1.5-2.0
nmol per mg protein following chromatography. These values agree very
well with the 17% of the non-heme iron not accounted for by ferritin
in microsomes. Lipid peroxidation dependent upon this nonvheme,
non'ferritin iron is SODvinsensitive, suggesting that it may be
directly reduced by microsomes, as is ADP-'Fe3+ (22). When iron
release experiments were conducted in the absence of purified ferritin
only very low rates were observed which were linear for only a short
period of time. This rate is likely to be reduction and release of
the nonvheme, nonvferritin iron and, in agreement with the lipid
peroxidation data, SOD did not inhibit this iron reduction and/or
release.

Chromatography of microsomes also resulted in a significant
decrease in catalase activity. Catalase has been shown to stimulate
lipid peroxidation dependent upon low concentrations of iron by
preventing H202 induced oxidation of ferrous iron (18). Therefore, it
was necessary to determine if the differences in rates of lipid
peroxidation between microsomal preparations was due to the .
differences in catalase activity. The addition of exogenous catalase
was found to have little effect in both microsomal preparations,
irrespective of the presence of ferritin. Previous work had
demonstrated that 10 units of catalase activity greatly stimulated
liposomal peroxidation (18,19), thus the catalase activity remaining

in microsomes is likely to be sufficient to prevent H202 accumulation.

 

227

These results suggest that the differences in the rates of lipid
peroxidation are not a result of the differences in catalase activity
but are attributable to the nonvheme, non-ferritin iron pool.

The results presented demonstrate that microsomes are capable of
releasing iron from ferritin and that ferritin is present in
microsomes prepared by the usual differential centrifugation technique.
While it is known that the microsomes generate only low amounts of 027
it is apparently produced in quantities sufficient to release ferritin
iron. It should be noted that similar results are obtained if
microsomes are washed with 10 mM EDTA, however the EDTA must then be
judiciously removed prior to use of the microsomes in lipid
peroxidation studies. The data indicate that microsomes also contain
a small amount of nonvheme iron which does not require 02? for
reduction. Our recent work indicates that both ferritin and the
unidentified, nonvheme iron pool are increased in microsomes following
ironvloading of rats (unpublished work). It will be of interest to
determine whether ferritin is associated with the endoplasmic
reticulum in vivo to further assess its potential to serve as a source

of iron for promotion of lipid peroxidation.

 

5.

10.

11.

12.

13.

LIST OF REFERENCES

P. Hochstein, K. Nordenbrand and L. Ernster. Evidence for the
involvement of iron in the ADPvactivated peroxidation of
lipids in microsomes and membranes. Biochem. Biophys. Res.
Comm. LA: 323 328 (196A). ‘ ‘ ‘

D. J. Kornburst and R. D. Mavis. Microsomal lipid peroxidation.
1. Characterization of the role of iron and NADPH. Mol. '
Pharmacol. 17: “00'407 (1980). ‘ '

S.D. Aust, L.A. Morehouse and C.E. Thomas. Role of metals in
oxygen radical reactions. J. Free Rad. Biol. Med. 1: 3-25
(1985). ' ' ' ' '

E.D. Wills. Mechanisms of lipid peroxide formation in tissues.
Role of metals and hematin proteins in the catalysis of the
oxidation of unsaturated fatty acids. Biochim. Biophys. Acta
98: 238-251 (1965). ' '

G.R. Bartlett. Phosphate compounds in rat erythrocytes and
reticulocytes. Biochem. Biophys. Res. Comm. 19; 1055'1062
(1976). - ' - - - -

S. Pollack, T. Campana and J. Weaver. Low molecular weight
iron in guinea pig reticulocytes. Amer. J. Hematol. lg: 75'8”
(1985). ‘ ' "

A. Jacobs. Ferritin: an interim review. In: Current Topics

 

 

in Hematology, Volume V (S. Piomelli and S. Yachnin, eds.) PP.
25'62, Alan R. Liss, New York (1985). ‘

P. M. Harrison. Ferritin: an ironvstorage molecule. Sem.
Hematol. 1n: 55'70 (1977) '

E.D. Wills. Lipid peroxide formation in microsomes. The role
of non-heme iron. Biochem. J. 113: 325'332 (1965).'

N.R. Montgomery, C. Clark and J.L. Holtzman. Iron species of
hepatic microsomes from control and phenobarbitalvtreated rats.
Arch. Biochem. Biophys. 160: 113'118 (197“).

K. S. Sargent and H. N. Munro. Association of ferritin with
liver cell membrane fractions. Exp. Cell Res. 9;: 15'22
(1975) ' ' '

E.C. Theil. Ferritin: structure, function, and regulation.
In:‘ Iron Binding Proteins Without Cofactors or Sulfur
Clusters (E. C. Theil, G. L. Eichorn and L. G. Marzilli, eds.)

pp. 1~38, Elsevier, New York (1983).

E.D. Wills. Mechanisms of lipid peroxide formation in animal
tissues. Biochem. J. 22: 667-676 (1966).

228

 

1h.

15.

16.

17.

18.

19.

20.

21.

22.

23.

2“.

25.

229

J.M.C. Gutteridge, B. Halliwell, A. Treffry, P.M. Harrison and
D. Blake. Effect of‘ferritinvcontaining fractions with
different iron loading on lipid peroxidation. Biochem. J.
209: 557:560 (1983). '

B. Rowley and G.D. Sweeney. Release of ferrous iron from
ferritin by liver microsomes: a possible role in the toxicity
of 2,3,7,8'tetrachlorodibenZOvpvdioxin. Can. J. Biochem. and
Cell Biol. gg: 1293*1300 (198M). ' ‘ ‘ '

R. Ulvik and I. Romslo. Studies on the mobilization of iron
from ferritin by isolated rat liver mitochondria. Biochim.
Biophys. Acta 588: 256~271 (1979).

R. J. Ulvik. Relevance of ferritinvbinding sites on isolated
mitochondria to the mobilization of iron from ferritin.
Biochim. Biophys. Acta 715: “2'51 (1982).

C.E. Thomas, L.A. Morehouse and S.D. Aust. Ferritin and
superoxidevdependent lipid peroxidation. J. Biol. Chem.

399: 3275-3280 (1985).

M. Saito, C. E. Thomas and S. D. Aust. Paraquat and
ferritinvdependent lipid peroxidation. J. Free Rad. Biol. Med.
1: 179-185 (1985L ' ' ' ‘ '

P. Biemond, H.G. VanEiJk, A.J.G. Swaak and J.F. Koster. Iron
mobilization from ferritin by superoxide derived from
stimulated polymorphonuclear leukocytes. J. Clin. Invest.

1;: 1576*1579 (1984).

R.L. Fong, P.B. McCay, J.L. Poyer, B.B. Keele and H. Misra.
Evidence that peroxidation of lysosomal membranes is initiated
by hydroxyl free radicals produced during flavin enzyme
activity. J. Biol. Chem. 238: 7792x7797 (1973).

L.A. Morehouse, C.E. Thomas and S.D. Aust. Superoxide
generation by NADPchytochrome PASO reductase: the effect of
iron chelators and the role of superoxide in microsomal lipid
peroxidation. Arch. Biochem. Biophys. 232: 366-377 (198").

J.M. McCord and I. Fridovich. Superoxide dismutase. An
enzymic function for erythrocuprein (hemocuprein). J. Biol.
Chem. 2““: 60u9'6055 (1969). ’ ‘

R. F. Beers, Jr. and I. W. Sizer. A spectrophotometric method
for measuring the breakdown of hydrogen peroxide by catalase.
J. Biol. Chem. 195:133'1h0 (1952L

T.C. Pederson, J.A. Buege and S.D. Aust. Microsomal electron
transport. The role of reduced nicotinamide adenine
dinucleotide phosphate-cytochrome 0 reductase in liver
microsomal lipid peroxidation. J. Biol. Chem. 238: 713h'71h1
(1973). - - ‘ - -

26.

27.

28.
29.
30.

31.

32.

33.

3M.
35.
36.

37.

230

T. Omura and R. Sato. The carbon monoxidevbinding pigment of
liver microsomes. II. Evidence for its hemOprotein nature.
J. Biol. Chem. 239: 2370*2378 (196R).

A. Horvat, J. Baxandall and O. Touster. The isolation of
lysosomes from Ehrlich Ascites tumor cells following
pretreatment of mice with Triton WR 1339 J. Cell Biol.
H2: N69vh79 (1969L ‘

J. H. Halliday. Immunoassay of ferritin in plasma. In:
Methods in Enzymology, Volume 8“ (J. Langone and J. Van
Vunakis, eds.) pp. 1H8'171, Academic Press, New York (1982).

T.S.L. Fan and P.S. Chu. An indirect enzyme-linked
immunosorbent assay for detection of aflatoxin B1 in corn and
peanut butter. J. Food Protec. £1: 263'266 (198”).

T. C. Pederson and S. D. Aust. Aminopyrine demethylase.
Kinetic evidence for multiple microsomal activities. Biochem.
Pharmacol. 19: 2221*2230 (1970).

M.W. Kennedy, N.R. Carpentier, P.P. Dymerski, S.M. Adams and
L. S. Kaminsky. Metabolism of monochlorobiphenylS'by hepatic
microsomal cytochrome PHSO. Biochem. Pharmacol. 22: 727'736
(1980). '

J.A. Buege and S.D. Aust. Microsomal lipid peroxidation. In:
Methods in Enzymology, Volume 52 (S. Fleischer and L. Packer,

 

eds.) pp. 302~310, Academic Press, New York (1978).

P.E. Brumby and V. Massey. Determination of nonvheme iron,
total iron, and COpper. In: Methods in Enzymology, Volume 10
(R. W. Estabrook and M. D. Pullman, eds.) pp. H63~H7N, Academic
Press, New York (1967).

 

J. R. Gillette, B. B. Brodie and B. M. LaDu. Oxidation of drugs
by liver microsomes: on the role of TPNH and oxygen. J.
Pharmacol. Exp. Ther. 119: 532-5N0 (1957L

L.A. Morehouse, M. Tien, J.R. Bucher and S.D. Aust. Effect of
hydrogen peroxide on the initiation of microsomal lipid
peroxidation. Biochem. Pharmacol. g2: 123-127 (1983).

C. Beauchamp and I. Fridovich. Superoxide dismutase: improved
assays and an assay applicable to acrylamide gels. Anal.
Biochem. L“: 276-287 (1971L ’

M. Tien, L.A. Morehouse, J.R. Bucher and S.D. Aust. The

multiple effects of ethylenediaminetetracetate in several
model lipid peroxidation systems. Arch. Biochem. Biophys.
gig: “SOVHSB (1982). ' '

38.

39.

231

A.L. Hunter and R.A. Neal. Inhibition of hepatic
mixed'function oxidase activity in vitro and in vivo by
various thionovsulfurvcontaining compounds. Biochem.
Pharmacol. 2“: 2199-2205 (1975)

C.M. Bartoli, A. Muller, E. Cadenas and H. Sies. Antioxidant
effect of diethyldithiocarbamate on lipid peroxidation
assessed by lowvlevel chemiluminescence and alkane production.
FEBS Lett. 16H: 371-37” (1983).

'

 

APPENDIX

Publications

 

Thomas, C. E., and S. D. Aust (1986), "Reductive release of iron from
ferritin by paraquat and related bipyridyls". J. Biol. Chem.,
In Press. ‘ ' '

Thomas, C. E., and S. D. Aust (1986), "Release of iron from ferritin by
cardiotoxic anthracycline antibiotics". Arch. Biochem. BiOphys.,
In Press. ' ' '

Thomas, C.E., and S.D. Aust (1986), "Free Radicals and Environmental
Toxins", Annals of Emerg. Med., In Press.

Thomas, C.E., and S.D. Aust (1986), "Role of metals in oxygen radical
reactions and oxidative stress", In: Handbook of Biomedicine of
Free Radicals and Antioxidants, (Miquel, J., ed. ), CRC Press,
Boca Raton, FL., In Press.

 

 

Thomas, C.E., L.A. Morehouse, and S.D. Aust (1985). "Ferritin and
superoxide-dependent lipid peroxidation", J. Biol. Chem. 260:
3275-3280. ‘ ‘ '

Thomas, C.E., and S.D. Aust (1985). "Rat liver microsomal NADPH-
dependent release of iron from ferritin and lipid peroxidation",
J. Free Rad. Biol. Med. 3: 293-300.

Saito, M., C.E. Thomas, and S.D. Aust (1985), "Paraquat and ferritin-
dependent lipid peroxidation", J. Free Rad. Biol. Med. 1:
179-185. ' ' ‘ ' ‘

Aust, S.D., L.A. Morehouse, and C. E. Thomas (1985), "Role of metals in
oxygen radical reactions", J. Free Rad. Biol. Med. 1: 3-25.

Aust, S.D., C.E. Thomas, L.A. Morehouse, M. Saito, and J.R. Bucher
(1985), "Active Oxygen and Toxicity", In: Biological Reactive
Intermediates III, Pergamon Press, Inc., New York, NY., In Press.

 

 

Aust, S.D., C.E. Thomas, L.A. Morehouse, and M. Saito (1985),
"Membrane Oxygen Radical Formation and Lipid Peroxidation", In:
Cell Membranes and Cancer, (Galeotti, T., A. Cittadini, G. Neri,
S. Papa, and L.A. Smets, eds.). Elsevier, Amsterdam, pp. 235—2uu.

 

Morehouse, L.A., C.E. Thomas, and S.D. Aust (198“), "Superoxide
generation by NADPH-cytochrome'PNSO reductase: The effect of
iron chelators and the role of superoxide in microsomal lipid
peroxidation", Arch. Biochem. Biophys. ggg: 366-377.

232

233

Abstracts

Thomas, C.E., and S.D. Aust (1986), "Radical release of iron from
ferritin", The Toxicologist 6: 152.

Thomas, C.E., and S.D. Aust (1985), "Release of ferritin iron by redox
active compounds", Conference on Radical Induced DNA Damage, East
Lansing, MI.

Thomas, C.E., M. Saito, and S.D. Aust (1985), "Superoxide mediated,
ferritin-dependent lipid peroxidation", Gordon Research
Conference, Ventura, CA.

Thomas, C.E., L.A. Morehouse, and S.D. Aust (1989), "Ferritin-
dependent microsomal and superoxide catalyzed lipid
peroxidation", Fed. Proc. 3;: 2029.

Thomas, C.E., L.A. Morehouse, and S.D. Aust (198M), "Ferritin and
superoxide-dependent lipid peroxidation", Great Lakes Workshop on
Oxygen Radicals in Medicine and Biology, East Lansing,’MI.

Thomas, C.E., L.A. Morehouse, and S.D. Aust (1989), "Superoxide-
dependent, ferritin catalyzed lipid peroxidation", Michgan
Regional Chapter of the Society of Toxicology, East Lansing, MI.

Saito, M., C.E. Thomas, L.A. Morehouse, and S.D. Aust (1989),
"Paraquat and ferritin-dependent lipid peroxidation", Great
Lakes Workshop on Oxygen Radicals in Medicine and Biology, East
Lansing, MI.

Morehouse, L.A., C.E. Thomas, and S.D. Aust (1983). "The effect of
EDTA-Fe3+ on NADPH cytochrome PNSO reductase-dependent superoxide
generation and lipid peroxidation", Fed. Proc. 32: 2062.

Morehouse, L.A., J.A. Bumpus, R. Voorman, C.E. Thomas, and S.D. Aust
(198R), "Superoxide production by reconstituted cytochrome
PNSO-dependent mixed function oxidases", Fed. Proc. 3;: 1N77.